SlideShare a Scribd company logo
1 of 6
Download to read offline
Session S1E
Recommender: Helping Viewers in their Choice for
   Educational Programs in Digital TV Context
                        Paulo Muniz de Ávila, Elaine Cecília Gatto, Sergio Donizetti Zorzo
                       pauloavila @pucpocos.com.br,elaine_gatto@dc.ufscar.br,zorzo@dc.ufscar.br



Abstract - Currently in Brazil, a fundamental change is       favorite program. In face of this situation, personalized
taking place in TV: the migration from analogue to            recommendation systems are necessary.
digital TV system. This change has two main                        Different from EPG functions which allow basic search,
implications: an increase in transmission capacity for        a personalized TV system can create a profile for each TV
new channels with the same bandwidth and the ability to       viewer and recommend programs that best match this
send applications with multiplexed audio-visual content.      profile, avoiding the search in many EPG options to find the
Brazilian government aims to exploit the transmission         favorite program. Elementary and secondary education
capacity for new channels offering programming created        schools and universities generally seek to explore this new
to distance learning and thereby promoting social             model offering personalized content to their students. In this
inclusion in the vast majority of the population. This        context, a recommendation system is able to analyze the
information overload demands mechanisms to help               profile of a group of students, suggesting the educational
students to browse and select what education programs         content that best suits the needs of the group.
are best suited to their current level. Personalized               To make the benefits (new channels, interactive
recommendation systems emerge as a solution to this           applications) offered by the digital system possible, the TV
problem, providing the viewer with educational                viewers with analogical system need new equipment called
programs relevant to his profile. In this paper we present    set-top box (STB). STB is a device which works connected
a     personalized    recommendation         system,    the   to the TV and converts the digital sign received from the
Recommender        consistent    with      the    reference   provider to audio/video that the analogical TV can exhibit.
implementation of the Brazilian digital TV system.            To have the advantages offered by the digital TV, the STB
Finally, we present the results obtained after using the      needs a software layer which connects the hardware to the
proposed system.                                              interactive applications called middleware. The DTV
                                                              Brazilian System middleware is Ginga [2,3]. It allows
Key-words - Personalization, Multimedia, Recommendation       declarative and procedural applications through its
System, Digital TV, Middleware Ginga.                         components Ginga-NCL [2] and Ginga-J [3]. Ginga-NCL
                                                              performs declarative application written in Nested Context
                     INTRODUCTION                             Language (NCL) while Ginga-J can perform procedural
     Digital television has created new services, products,   application based on JavaTM known as Xlets [4].
contents, channels and business models. The Brazilian              This paper proposes an extension to Ginga middleware
Digital TV System allows high quality audio and video, as     through implementation of a new module incorporated to
well as interactivity, creating different contents for users. Ginga Common Core called Recommender. The
There are two main implications with Brazil Digital TV        Recommender module is responsible for gathering, storing,
System: the increase of the number of channels being          processing and recommending TV education programs. To
broadcasted with the same bandwidth and the possibility of    develop the Recommender module, Ginga-NCL middleware
sending multiplexed applications with the audio-visual        developed by PUC-RIO (Pontifical Catholic University of
content. As new channels emerge due to the transmission       Rio de Janeiro) was used, implemented in C/C++ language
increase, it is necessary to create ways that allow the TV    with source code available under GPLv2 license and
viewers to search among these channels.                       according with the patterns defined by the Brazilian system
     The Electronic Program Guide (EPG) helps the TV          digital television [4].
viewers. However, as new channels are available, an                      TVDI IN BRAZIL AND EDUCATION
information overload is unavoidable making the EPG system
inappropriate. In Shangai [1], a big city in China, the TV One of the reasons to implement TVDi in the national
operators provide different services (in the analogical territory is its potential to social inclusion. In Brazil, in many
system, channels), and this number has been increasing at a cases, the open TV is the only source of information for
20% rate per year. Thus, the traditional EPG system became people who do not frequently read newspaper, magazine or
unattractive because it takes too long for the viewers to any other kind of printed media. If we consider that the
search among hundreds of options available to find their access to written information is low and that the information
978-1-4244-6262-9/10/$26.00 ©2010 IEEE                                        October 27 - 30, 2010, Washington, DC
                                    40th ASEE/IEEE Frontiers in Education Conference
                                                           S1E-1
Session S1E
transmitted through TV newscasts is the biggest link               features anymore due particularly to the financial survival of
between the world and the daily routine of Brazilian people,       theses broadcastings. It is possible to note, according to
we have many reasons not to ignore the reach power of this         legislation, that the programming only admits transmission
technology. If it is correctly and consciously explored, with      of programs with educative-cultural purposes. However,
the help of interactive resources, TVDi can represent a            there is the option to recreational, informative or sport
powerful tool to have access to differentiated educational         programs considered educative-cultural since they present
knowledge at the same time it can include Brazilian citizens       instructive elements or educative-cultural focus identified in
digitally excluded nowadays. Thus, it can be said that in          its presentation.
Brazil, the access to the Internet is low and high-class people    Digital TV implantation in Brazil has been advancing. Some
are those who have more access to it and participate               obstacles – among them the situation of commercial
somehow in the educational scenario. The low number of             broadcastings, political interests, influences (and models) of
personal computers and the high number of TV sets in               digital television international systems, legislation ruling the
Brazilian houses defend the efforts to use all TVDi potential      radio broadcasting – still prevent its complete operation, but
in issues in the educational extent. If public policies are well   when it is defined, a social participation never seen before in
structured, TVDi can reinforce a new educational paradigm,         other historical moments can take place in Brazil, ensuring
allowing the entire population to have access to Internet          access to information and culture. [6]
resources, video, images, sounds, interactivity to introduce
new knowledge, entertainment, education, leisure, services.                            RELATED WORKS
It can allow the unlimited access to written and audiovisual       There are several recommendation systems for DTV (Digital
information. As the great part of Brazilian population has a       Television) designed to offer a distinct personalization
limited access to information and Internet, and considering        service and to help TV viewers to deal with the great
the fact that the TV is the durable good which is in almost all    quantity of TV programs. Some systems related to the
Brazilian houses, we can consider the TVDi a way to                current work are presented here.
significantly change the perspective of Brazilian distance         The AIMED system proposed by [7], presents a
learning. Even knowing that the TVDi inclusion in Brazil           recommendation mechanism that considers some TV viewer
will not solve the social inclusion problem, it is certain that    characteristics as activities, interests, mood, TV use
all its power can improve the digital inclusion, for it will       background and demographic information. These data are
ensure the information access, services and education to           inserted in a neural network model that infers the viewers’
people with low purchasing power. [5]                              preferences about the programs. Unlike the work proposed
          EDUCATIVE BROADCASTING IN BRAZIL                         in this paper, which uses the implicit data collection, in the
                                                                   AIMED system, the data are collected and the system is set
According to the Communication Department, educative               trough questionnaires. This approach is doubtful, mainly
broadcasting is the Sound Broadcasting Service (radio) or          when limitations imposed to data input in a DTV system are
Sounds and Images Services (TV) intended for the                   considered.
transmission of educative-cultural programs which, besides         In [8] a method to discover models of multiuser environment
performing together with teaching systems of any level or          in intelligent houses based on users’ implicit interactions is
modality, aims the basic and higher education, the                 presented. This method stores information in logs. So, the
permanent education and the professional education, besides        logs can be used by a recommendation system in order to
comprehending educational, cultural, pedagogical and               decrease effort and adapt the content for each TV viewer as
professional orientation activities. The execution of              well as for multiuser situations. Evaluating the TV viewers’
broadcasting services with exclusively educative purposes is       background of 20 families, it was possible to see that the
granted to legal entities with internal public right, including    accuracy of the proposed model was similar to an explicit
universities, which will be given the preference to obtain the     system. This shows that collecting the data in an implicit
grant, and foundations privately established and others            way is as efficient as the explicit approach. In this system,
Brazilian universities.                                            the user has to identify himself in an explicit way, using the
The first educative broadcasting station, the University TV        remote control. Unlike this system, the proposal in this paper
of Pernambuco pertaining to the Education Department, was          aims at promoting services to the recommendation systems
on TV in 1967. Until 1980´s, educative TV broadcasting in          for a totally implicit multiuser environment.
Brazil gave priority to essentially educative programs and in      In [9], a program recommendation strategy for multiple TV
1997, the Brazilian Association of Public, Educative and           viewers is proposed based on the combination of the
Cultural Broadcasting (ABEPEC) was created. In 1999, the           viewer’s profile. The research analyzed three strategies to
participant broadcastings created the RPTV (Public TV              perform the content recommendation and provided the
Network) which aims at establishing a common and                   choice of the strategy based on the profile combination. The
mandatory programming guide to the associated                      results proved that the TV viewers’ profile combination can
broadcastings. Today, the programming is different from            reflect properly in the preferences of the majority of the
that one in the beginning of educative broadcasting                members in a group. The proposal in this paper uses an
transmissions, that is, it does not have the strict educative      approach similar to a multiuser environment, however,
978-1-4244-6262-9/10/$26.00 ©2010 IEEE                                   October 27 - 30, 2010, Washington, DC
                               40th ASEE/IEEE Frontiers in Education Conference
                                                   S1E-2
Session S1E
besides the profile combination, the time and day of the          a set of modules responsible for the data processing,
week are also considered.                                         information filtering in the transport stream. It is the
In [1] a personalized TV system is proposed loaded in the         architecture core; Stack protocol layer responsible for
STB compatible with the Multimidia Home Plataform                 supporting many communication protocols like HTTP, RTP
(MHP) model of the digital television European pattern.           and TS.
According to the authors, the system was implemented in a
commercial solution of the MHP middleware, and for that,
implemented alterations and inclusions of new modules in
this middleware. Offering recommendation in this system
requires two important information that must be available:
programs description and the viewer visualization behavior.
The description of the programs is obtained by
demultiplexing and decoding the information in the EIT
(Event Information Table) table. EIT is the table used to
transport specific information about programs, such as: start
time, duration and description of programs in digital
television environments. The viewing behavior is collected
monitoring the user action with the STB and the later
persistence of this information in the STB. The work of [1]
is similar to the work proposed in this paper. The implicit
collection of data, along with the inclusion of a new module       FIGURE 1 – GINGA MIDDLEWARE ARCHITECTURE (ADJUSTED
in the middleware architecture, is an example of this                        WITH THE RECOMMENDATION SYSTEM)
similarity.
In [10], the Personalized Electronic Program Guide is             The proposed system extends the Ginga middleware
considered a possible solution to the information overload        functionalities including new services in the Ginga Common
problem, mentioned in the beginning of this work. The             Core layer. The Recommender module is the main part of
authors compared the use of explicit and implicit profile and     the recommendation system and it is inserted in the
proved that the indicators of implicit interests are similar to   Common Core layer of Ginga-NCL architecture. The
the indicators of explicit interests. The approach to find out    Recommender module is divided in two parts. The first one
the user’s profile in an implicit way is adopted in this work     describes the components integrated to the source code of
and it is about an efficient mechanism in the context of          the middleware such as Local Agent, Schedule Agent, Filter
television environment, where the information input is            Agent and Data Agent. The second part describes the new
performed through remote control, a device that was not           component added to the STB: Sqlite [13], a C library which
designed to this purpose.                                         implements an attached relational database. Figure 2
In [11], the AVATAR recommendation system is presented,           presents the Recommender module architecture.
compatible to the European MHP middleware. The authors            I.   Implemented Modules
propose a new approach, where the recommendation system
is distributed by broadcast service providers, as well as an      This subsection describes the modules added to the Ginga-
interactive application. According to the authors, this           NCL middleware source code and the extensions
approach allows the user to choose among different                implemented to provide a better connection between
recommendation systems, what is not possible when we              middleware and the recommendation system.
have an STB with a recommendation system installed in             Local Agent is the module responsible for constant
plant. The AVATAR system uses the approach of implicit            monitoring of the remote control. Any interaction between
collection of user profile and proposes modifications in the      the viewer and the control is detected and stored in the
MHP middleware to include the monitoring method. The              database. The Local Agent is essential for the
Naïve Bayes [12] is used as a classification algorithm and        recommendation system that uses implicit approach to
one of the main reasons for that is the low use of STB            perform the profile.
resources.
                                                                  Scheduler Agent is the module responsible for periodically
                  SYSTEM OVERVIEW                                 request the data mining. Data mining is a process that
                                                                  demands time and processing, making its execution
The recommendation system proposed in this paper is based
                                                                  impracticable every time the viewer requests a
on Ginga middleware. As mentioned before, the version
                                                                  recommendation. Scheduler Agent module guarantees a new
used was the open source version of Ginga-NCL
                                                                  processing every 24 hours preferably at night, when the STB
middleware. Figure 1 presents its architecture consisting of
                                                                  is in standby.
three layers:
Resident applications responsible for the exhibition
(frequently called presentation layer); Ginga Common Core,
978-1-4244-6262-9/10/$26.00 ©2010 IEEE                                   October 27 - 30, 2010, Washington, DC
                               40th ASEE/IEEE Frontiers in Education Conference
                                                   S1E-3
Session S1E
                                                                items. For example, the system can be used to create a top-
                                                                10 question topic; the students would classify extra material
                                                                with a grade and the best extra materials would be
                                                                recommended. It would be also possible to have a top-10
                                                                favorite and a top-10 best students. Moreover, the system
                                                                could also provide a way to look for old content interesting
                                                                for the user to improve what is being studied at that moment.

                                                                               METHODOLOGY AND TESTS
                                                                     User history and EPG data are necessary to perform the
                                                                tests. These data were provided by IBOPE (Brazilian
                                                                Institute of Public Opinion and Statistics) [14] through a
                                                                treatment process almost entirely manual in order to be in
                                                                accordance to the standard format which must be used in the
                                                                Brazilian digital TV system and also in the tests.
                                                                     Many technologies have been arising with the aim at
                                                                identifying behavior standards and its application in the
     FIGURE 2 – RECOMMENDER MODULE ARCHITECTURE                 personalization. The recommendation systems operation is
                                                                found on these techniques and the most used are the
                                                                Collaborative Filtering and Content-Based Filtering which
Mining Agent is the module that accesses the information in     includes several algorithms for each one.                    A
the viewer’s behavior background and the programming data       recommendation system can use only one technique or two
from the EIT and SDT tables stored in cache to perform the      together, becoming a hybrid system.
data mining. In order to process the data mining, the Mining         In order to study, analyze and choose an algorithm to be
module has direct access to the database and recovers the       used in Technical module, some information filtering
TV viewer’s behavior background. From the point of view         algorithms were tested. The tests were performed in three
of the system performance, this communication between           steps. In the first step, tests were performed with Apriori
mining module and user database is important. Without this      algorithm. In the second step, the forecast method was used,
communication, it would be necessary to implement a new         applying Cosine as measure of similarity. The third step was
module responsible for recover the database information and     to compare the results and the operation with both
then make such data available to the mining algorithm. The      algorithms, analyzing the facilities and difficulties,
second data set necessary to make possible the data mining      especially for the implementation.
is the program guide. The program guide is composed by               The association techniques algorithms identify
information sent by providers through EIT and SDT tables.       associations between the data registers which are related in
These tables are stored in cache and are available to be        some way. The basic premise finds elements which imply
recovered and processed by the Mining module. Ginga-NCL         the presence of others in a same operation aiming at
Middleware does not implement storage mechanism in cache        determining which are related. The association rules
of EIT and SDT tables. This functionality was implemented       interconnect objects trying to show characteristics and
by the Recommender system.                                      tendencies. The association discoveries present trivial and
                                                                non trivial association. The data was adapted in order to be
Filter Agent & Data Agent The raw data returned by the          used in Apriori algorithm, that is, it was submitted to a pre-
Mining Agent module need to be filtered and later stored in     processing phase. The user history was created from IBOPE
the viewer’s database. The Filter Agent and Data Agent          data. For the implementation, it is not necessary that the data
modules are responsible for this function. The Filter Agent     go through adjustments, as it will be collected in the correct
module receives the data from the mining provided by the        format to be used. The results were satisfactory verifying
Mining Agent and eliminates any information that is not         that Apriori can be applied to the system for it can be
important keeping only those which are relevant to the          adapted to the system needs. [15, 16]
recommendation system such as the name of the program,               The Cosine is a similarity measure, a forecast method
time, date, service provider and the name of the service. The   which calculates the similarity between items and users,
Data Agent module receives the recommendations and stores       consults similar items to a given item and matches item
them in the viewer’s database.                                  content and user profile. The data also had to be adjusted to
                                                                be used with Cosine. Database in sqlite was used with the
If there were many educative programs on open TV, it            EPG and the user history. From these two tables, it was
would be very useful to recommend other educative               possible to derive two more, one with the profile of the
programs. However, “educative” is one of the many TV            program watched by the user and other with the profile of
program categories. The system can be used inside a             genres. It was necessary that the EPG passed through a
distance learning system to recommend several types of          modification which should also occur in the implementation.

978-1-4244-6262-9/10/$26.00 ©2010 IEEE                                   October 27 - 30, 2010, Washington, DC
                               40th ASEE/IEEE Frontiers in Education Conference
                                                   S1E-4
Session S1E
A new table was created, identical to the EPG table, but
added with fields containing the genres names. According to
the adjustment of the program in the genres, these fields
were populated with 0 or 1, becoming a matrix. From these
tables it was possible to find the Cosine for the programs and
genres, the profile and what could be recommended to the
user. The results from the Cosine were also satisfactory
confirming that this technique can be applied to the system
for it can be adjusted to the system needs. [17, 18]

                          ANALYSIS
     During the tests, it was possible to note some
particularities. Our system recommends contents based on
the programs genres and our analyses were performed
according to this standard. With Apriori algorithm, the data
are collected in the correct format to be used. For the Cosine,      FIGURE 3. ACCURACY OF THE RECOMMENDATION SYSTEM
the EPG needs to be changed to a matrix before starting the
process of discovering profiles and recommendations.                    Figure 3 presents the results obtained after 4 weeks of
     In a desktop, the feedback of the Cosine calculation is       monitoring considering the best value obtained among the 8
faster in relation to the feedback of Apriori association rules.   schools analyzed. It is clear that on the first weeks, as the
However, further studies about these algorithms processing         collected data were few, Apriori algorithm did not extract
in these devices are still being performed. Apriori is able to     relevant information from the preferences of the group. With
discover the profile from the standards, but to select the         the data increase in the visualization background on the third
programs to be recommended, another technique must be              and fourth week, the algorithm obtained better results and
used and the Cosine can find both the profile and the              the index of recommendation acceptance increased.
recommendations.
     The Cosine cannot discover these characteristics, but
reaches our goal. In order to discover behaviors similar to
the association rules, it is necessary to consult the databank.
Apriori output must be operated in order to give the correct
user profile, that is, the rules must be understood, and that is
very hard concerning implementation. The Cosine output is
clearer; the result straightly reaches intended goal, allowing
the output to be used without the need of a post-treatment.
     Regarding the input, there is no need of treatment for
Apriori, since all data will be used as they are collected.
However, for the Cosine, whenever the EPG is updated, the
table containing the EPG matrix must be changed according
to the new EPG, becoming something hard to work. The
profile of the genres founded by both algorithms is similar.

                           RESULTS
In order to measure the evolution of the recommendation            FIGURE 4 ACCURACY OF THE RECOMMENDATION SYSTEM PER
                                                                                          SCHOOL
offered to the students viewer, the following formula was
applied:
                                                                       Figure 4 presents the accuracy per school. The main
                    Ef = (α / β) 100                 (1)           characteristic of the schools is the socioeconomics difference
                                                                   among them. The conclusion is that Apriori algorithm had a
    Where Ef is the efficacy of the recommendation system,         good performance unrestricted to the students’
ranging from 0 to 100, α is the recommendation number              ´socioeconomic profile.
accepted by the students viewers and β is the number of
recommendation presented. In order to monitor these data
provided by IBOPE were used. The validation adopted an
accuracy formula presented in (1).




978-1-4244-6262-9/10/$26.00 ©2010 IEEE                                   October 27 - 30, 2010, Washington, DC
                               40th ASEE/IEEE Frontiers in Education Conference
                                                   S1E-5
Session S1E
                                                                             [3]    Souza Filho, G. L., Leite, L. E. C., Batista, C. E. C. F. 2007, “Ginga-J:
                                                                                    The Procedural Middleware for the Brazilian Digital TV System.”,
                                                                                    Journal of the Brazilian Computer Society, v. 12, n. 4, p. 47-56,
                                                                                    March 2007.
                                                                             [4]     Ginga-NCL Virtual STB, (March 2009), Available at:
                                                                                    http://www.ncl.org.br/ferramentas/index_30.html.
                                                                             [5]    Silva, Dirceu et al. Possibilidades educativas e de inclusão social e
                                                                                    digital com a TVDi: uma breve análise do cenário brasileiro.
                                                                                    Universidade Estadual de Campinas, Brasil. Available in:
                                                                                    http://www.google.com.br/url?sa=t&source=web&ct=res&cd=1&ved
                                                                                    =0CAYQFjAA&url=http%3A%2F%2Fwww.rieoei.org%2Fdeloslecto
                                                                                    res%2F2907Veraszto.pdf&ei=4ar5S-
                                                                                    eeM8imuAeA1PG9Dg&usg=AFQjCNHaebMawhyd-
                                                                                    xbrBsw0JjSHbTr7LQ&sig2=VGSGkTroJZu8t3NYiHA9_A. Acess in
                                                                                    2010-05-20.
                                                                             [6]    Fort, Mônica Cristine. Televisão + Educação = Televisão Educativa.
                                                                                    Available in:
                                                                                    http://www.google.com.br/url?sa=t&source=web&ct=res&cd=4&ved
                                                                                    =0CBUQFjAD&url=http%3A%2F%2Fencipecom.metodista.br%2Fm
                                                                                    ediawiki%2Fimages%2Fb%2Fbf%2FGT10_-
                                                                                    _008.pdf&ei=N8n5S5TaDYmHuAfp5Py9Dg&usg=AFQjCNFLgg9ng
                                                                                    2elo7UJAjo9dpf8-3I4hg&sig2=RrpI9k-iotN31hCBrc6yNw. Acess in
              FIGURE 5 – RECOMMENDERTV SYSTEM                                       2010-05-20.
                                                                             [7]    S. H. Hsu, M. H. Wen, H. C. Lin, C. C. Lee, C. H. Lee. 2007.
    Figure 5 shows Recommender system. The application                              “AIMED – A personalized TV Recommendation System” in Proc
                                                                                    2007 Interactive TV: A Shared Experience. 5th European Conference,
used as front-end is written in NCL and allows the students                         EuroITV 2007, Amsterdam, the Netherlands.
to search the recommendation list selecting the education                    [8]    Vildjiounaite, E., Kyllonen, V., Hannula, T. and Alahuhta, P. 2008.
program.                                                                            Unobtrusive Dynamic Modelling of TV Program Preferences. In
                                                                                    Proceedings of the Changing Television Environments, 6th European
                            CONCLUSION                                              Conference, EuroITV 2008, pages 82-91.
                                                                             [9]    Zhiwen, Y., Xingshe, Z., Yanbin, H. and Jianhua, G. 2006. TV
     With the appearance of digital TV, a variety of new                            program recommendation for multiple viewers based on user profile
services (in the analogical system, channels) will be                               merging. In Proceedings of the User Modeling and User-Adapted
available. This information overload requires the                                   Interaction, pages 63-82. Publishing Springer Netherlands.
implementation of new mechanisms to offer facilities to the                  [10]   O’Sullivan, D., Smyth, B., Wilson, D. C., McDonald, K. and
                                                                                    Smeaton, A. 2004. Interactive Television Personalization: From
students looking for their education programs. These new                            Guides to Programs. Personalized Digital Television: Targeting
mechanisms suggesting the viewers programs are known as                             Programs to Individual Viewers. L. Ardissono, A. .Kobsa and M.
recommendation systems. A recommendation system                                     Maybury editors, pages 73-91, Kluwer Academic Publishers
compatible with Ginga middleware is presented in this paper                  [11]   Blanco-Fernandez, Y., Pazos-Arias, J., Gil-Solla, A., Ramos-Cabrer,
and it is implemented according to the standards of the                             M.,Lopes-Nores, M., Barragans-Martinez, B. 2005. AVATAR: a
                                                                                    Multi-agent        TV         Recommender            System        Using
digital television Brazilian system. The recommendation                             MHP Applications. In: IEEE International Conference on E-
system was modeled considering the current characteristics                          Technology, E-Commerce and E-Service (EEE '05), pp. 660-665.
of the television, and this model can be adjusted to other                   [12]   Wu, X., Kumar, V., Ross Quinlan, J., Ghosh, J., Yang, Q., Motoda,
standards and also to new portable devices which will be on                         H., McLachlan, G. J., Ng, A., Liu, B., Yu, P. S., Zhou, Z., Steinbach,
                                                                                    M., Hand, D. J., and Steinberg, D. 2007. Top 10 algorithms in data
the market. At last, future works can include algorithms of                         mining. Knowl. Inf. Syst. 14, 1 (Dec. 2007), 1-37. DOI=
collaborative filtering and also a new architecture using                           http://dx.doi.org/10.1007/s10115-007-0114-2
client-server, providing and offering other kinds of                         [13]   Sqlite, (March 2010) Available at: http://www.sqlite.org/.
personalization services for the users.                                      [14]   “IBOPE”. Available in: http://www.ibope.com.br. Access in
                                                                                    December 2009.
                       ACKNOWLEDGMENT                                        [15]   Witten, I. H.; Frank, E. Data Mining: Practical Machine Learning
                                                                                    Tools and Techniques, 2nd Edition, Morgan Kaufmann, 525 pages,
     We thank IBOPE for providing real data about the                               June 2005.
electronic program guide and also the viewer’s behavior data                 [16]   Gatto, Elaine C.; Zorzo, Sergio D. “Sistema de Recomendação para
from March, 05, 2009 to March, 19, 2009.                                            TVDPI,” in 8th International Information and Telecommunication
                                                                                    Technologies Symposium. Florianópolis, Santa Catarina, Brasil. 09-
                            REFERENCES                                              11/12/2009.
                                                                             [17]   Torres, Roberto. “Personalização na Internet.” Novatec Editora. 2004.
[1]   H. Zhang, S. Zheng and J. Yuan. 2005, “A Personalized TV Guide                158p.
      System Compliant with MHP”, IEEE Transactions on Consumer
                                                                             [18]   Gatto, Elaine C.; Zorzo, Sergio D. Application of recommendation
      Electronics, pages 731-737, Vol. 51, No. 2, MAY 2005.
                                                                                    techniques for Brazilian Portable Interactive Digital TV. In: IWSSIP
[2]   L.F.G. Soares, R.F. Rodrigues, M.F. Moreno. 2007, “Ginga-NCL: The             2010 - 17th International Conference on Systems, Signals and Image
      declarative Environment of the Brazilian Digital TV System”, Journal          Processing. June 17-19, 2010, Rio de Janeiro, Brazil.
      of the Brazilian Computer Society. V.12, n.4, p.37-46, March 2007.



978-1-4244-6262-9/10/$26.00 ©2010 IEEE                                   October 27 - 30, 2010, Washington, DC
                               40th ASEE/IEEE Frontiers in Education Conference
                                                   S1E-6

More Related Content

What's hot

Television Network Strategy Research Paper
Television Network Strategy Research PaperTelevision Network Strategy Research Paper
Television Network Strategy Research PaperOrlando O'Neill
 
Crc broadcast technical panel summary november 2010
Crc broadcast technical panel summary november 2010Crc broadcast technical panel summary november 2010
Crc broadcast technical panel summary november 2010CRCPanel
 
Roadmap strategies chart draft 2.0
Roadmap strategies chart draft 2.0Roadmap strategies chart draft 2.0
Roadmap strategies chart draft 2.0Linda Shum
 
Mobiles4Development Case Studies
Mobiles4Development Case StudiesMobiles4Development Case Studies
Mobiles4Development Case Studiessegughana
 
2012 02-17 rcbf event - combined presentations - all slides
2012 02-17 rcbf event - combined presentations - all slides2012 02-17 rcbf event - combined presentations - all slides
2012 02-17 rcbf event - combined presentations - all slidesGlenn Peacey
 
EDHEC_Master project_Marine NOGIER_without appendice
EDHEC_Master project_Marine NOGIER_without appendiceEDHEC_Master project_Marine NOGIER_without appendice
EDHEC_Master project_Marine NOGIER_without appendiceMarine Nogier
 
How Can You Enhance Your Skills and Library Practice at Your ...
How Can You Enhance Your Skills and Library Practice at Your ...How Can You Enhance Your Skills and Library Practice at Your ...
How Can You Enhance Your Skills and Library Practice at Your ...Videoguy
 
Unicef Report Mobiles For Development
Unicef Report Mobiles For DevelopmentUnicef Report Mobiles For Development
Unicef Report Mobiles For Developmentsegughana
 
Case study of use of IVR systems to empower the visually impaired
Case study of use of IVR systems to empower the visually impairedCase study of use of IVR systems to empower the visually impaired
Case study of use of IVR systems to empower the visually impairedAshish Tandon
 
Digital agenda for cyprus
Digital agenda for cyprusDigital agenda for cyprus
Digital agenda for cyprusAnima Slides
 

What's hot (11)

Television Network Strategy Research Paper
Television Network Strategy Research PaperTelevision Network Strategy Research Paper
Television Network Strategy Research Paper
 
Crc broadcast technical panel summary november 2010
Crc broadcast technical panel summary november 2010Crc broadcast technical panel summary november 2010
Crc broadcast technical panel summary november 2010
 
Roadmap strategies chart draft 2.0
Roadmap strategies chart draft 2.0Roadmap strategies chart draft 2.0
Roadmap strategies chart draft 2.0
 
Mobiles4Development Case Studies
Mobiles4Development Case StudiesMobiles4Development Case Studies
Mobiles4Development Case Studies
 
2012 02-17 rcbf event - combined presentations - all slides
2012 02-17 rcbf event - combined presentations - all slides2012 02-17 rcbf event - combined presentations - all slides
2012 02-17 rcbf event - combined presentations - all slides
 
EDHEC_Master project_Marine NOGIER_without appendice
EDHEC_Master project_Marine NOGIER_without appendiceEDHEC_Master project_Marine NOGIER_without appendice
EDHEC_Master project_Marine NOGIER_without appendice
 
MDG tracking
MDG trackingMDG tracking
MDG tracking
 
How Can You Enhance Your Skills and Library Practice at Your ...
How Can You Enhance Your Skills and Library Practice at Your ...How Can You Enhance Your Skills and Library Practice at Your ...
How Can You Enhance Your Skills and Library Practice at Your ...
 
Unicef Report Mobiles For Development
Unicef Report Mobiles For DevelopmentUnicef Report Mobiles For Development
Unicef Report Mobiles For Development
 
Case study of use of IVR systems to empower the visually impaired
Case study of use of IVR systems to empower the visually impairedCase study of use of IVR systems to empower the visually impaired
Case study of use of IVR systems to empower the visually impaired
 
Digital agenda for cyprus
Digital agenda for cyprusDigital agenda for cyprus
Digital agenda for cyprus
 

Viewers also liked

Viewers also liked (10)

H6 het parlement
H6 het parlementH6 het parlement
H6 het parlement
 
H9 politieke besluitvorming
H9 politieke besluitvormingH9 politieke besluitvorming
H9 politieke besluitvorming
 
Rol media maatschappelijk vraagstuk
Rol media maatschappelijk vraagstukRol media maatschappelijk vraagstuk
Rol media maatschappelijk vraagstuk
 
H10 de toekomst van de Nederlandse politiek
H10 de toekomst van de Nederlandse politiekH10 de toekomst van de Nederlandse politiek
H10 de toekomst van de Nederlandse politiek
 
H7 gemeente en provincie
H7 gemeente en provincieH7 gemeente en provincie
H7 gemeente en provincie
 
Van verkiezing tot kabinet
Van verkiezing tot kabinetVan verkiezing tot kabinet
Van verkiezing tot kabinet
 
H5 kabinet en regering deel 2
H5 kabinet en regering deel 2H5 kabinet en regering deel 2
H5 kabinet en regering deel 2
 
H5 kabinet en regering deel 1
H5 kabinet en regering deel 1H5 kabinet en regering deel 1
H5 kabinet en regering deel 1
 
H8 Nederland en Europa
H8 Nederland en EuropaH8 Nederland en Europa
H8 Nederland en Europa
 
Parlementaire democratie h6
Parlementaire democratie   h6Parlementaire democratie   h6
Parlementaire democratie h6
 

Similar to Fie recommender helping viewers in their choice for educational programs in digital tv context

Recommender helping viewers in their choice for educational programs in digit...
Recommender helping viewers in their choice for educational programs in digit...Recommender helping viewers in their choice for educational programs in digit...
Recommender helping viewers in their choice for educational programs in digit...Elaine Cecília Gatto
 
Using content-based filtering in a system of recommendation in the context of...
Using content-based filtering in a system of recommendation in the context of...Using content-based filtering in a system of recommendation in the context of...
Using content-based filtering in a system of recommendation in the context of...Elaine Cecília Gatto
 
BIPODITVR: brazilian interactive portable digital tv recommendation system
BIPODITVR: brazilian interactive portable digital tv recommendation systemBIPODITVR: brazilian interactive portable digital tv recommendation system
BIPODITVR: brazilian interactive portable digital tv recommendation systemElaine Cecília Gatto
 
T-learning for social inclusion
T-learning for social inclusionT-learning for social inclusion
T-learning for social inclusioneLearning Papers
 
Application of recommendation techniques for brazilian portable interactive d...
Application of recommendation techniques for brazilian portable interactive d...Application of recommendation techniques for brazilian portable interactive d...
Application of recommendation techniques for brazilian portable interactive d...Elaine Cecília Gatto
 
BEACON: Can Digital Terrestrial Television lead to an inclusive society?
BEACON: Can Digital Terrestrial Television lead to an inclusive society?BEACON: Can Digital Terrestrial Television lead to an inclusive society?
BEACON: Can Digital Terrestrial Television lead to an inclusive society?hbijnens
 
Iwssip application of recommendation techniques for brazilian portable inte...
Iwssip   application of recommendation techniques for brazilian portable inte...Iwssip   application of recommendation techniques for brazilian portable inte...
Iwssip application of recommendation techniques for brazilian portable inte...Elaine Cecília Gatto
 
Cable Television Video-On-Demand for Learner-Centered Instruction:on-manuscript
Cable Television Video-On-Demand for Learner-Centered Instruction:on-manuscriptCable Television Video-On-Demand for Learner-Centered Instruction:on-manuscript
Cable Television Video-On-Demand for Learner-Centered Instruction:on-manuscriptCJ Cornell
 
Business Models for Web TV - Research Report
Business Models for Web TV - Research ReportBusiness Models for Web TV - Research Report
Business Models for Web TV - Research ReportAlessandro Masi
 
DTC meeting at the ITU, UN by Dr Danica Radovanovic
DTC meeting at the ITU, UN by Dr Danica RadovanovicDTC meeting at the ITU, UN by Dr Danica Radovanovic
DTC meeting at the ITU, UN by Dr Danica RadovanovicDr Danica Radovanovic
 
MEGATRENDS SM DIPATANE.pdf
MEGATRENDS SM DIPATANE.pdfMEGATRENDS SM DIPATANE.pdf
MEGATRENDS SM DIPATANE.pdfSamDipatane1
 
Web-TV an Innovative & Effective Digital Communication Tool for Think Tanks
Web-TV an Innovative & Effective Digital Communication Tool for Think TanksWeb-TV an Innovative & Effective Digital Communication Tool for Think Tanks
Web-TV an Innovative & Effective Digital Communication Tool for Think TanksRaja Taimur Hassan
 
Zahid Hussain - Internet Tv Aug 2008 Poland
Zahid Hussain - Internet Tv Aug 2008 PolandZahid Hussain - Internet Tv Aug 2008 Poland
Zahid Hussain - Internet Tv Aug 2008 Polandguest4d4d00
 
IAMCR 2004 - The Consequence of Digital TV for the Brazilian Media Professional
IAMCR 2004 - The Consequence of Digital TV for the Brazilian Media ProfessionalIAMCR 2004 - The Consequence of Digital TV for the Brazilian Media Professional
IAMCR 2004 - The Consequence of Digital TV for the Brazilian Media Professionalsaramuller
 
UNESCO-Pearson Initiative: Digital Inclusion Landscape Review
UNESCO-Pearson Initiative: Digital Inclusion Landscape ReviewUNESCO-Pearson Initiative: Digital Inclusion Landscape Review
UNESCO-Pearson Initiative: Digital Inclusion Landscape ReviewSteve Vosloo
 
Department of Health digital comms strategy 2011
Department of Health digital comms strategy 2011Department of Health digital comms strategy 2011
Department of Health digital comms strategy 2011Tim Lloyd
 
Innovative business model based on DVB-H and 3G synergy
Innovative business model based on DVB-H and 3G synergyInnovative business model based on DVB-H and 3G synergy
Innovative business model based on DVB-H and 3G synergyMichal Marcinik
 
AN ANALYSIS OF FACTORS AFFECTING THE ADOPTION OF DIGITAL TERRESTRIAL TELEVISI...
AN ANALYSIS OF FACTORS AFFECTING THE ADOPTION OF DIGITAL TERRESTRIAL TELEVISI...AN ANALYSIS OF FACTORS AFFECTING THE ADOPTION OF DIGITAL TERRESTRIAL TELEVISI...
AN ANALYSIS OF FACTORS AFFECTING THE ADOPTION OF DIGITAL TERRESTRIAL TELEVISI...ijmpict
 

Similar to Fie recommender helping viewers in their choice for educational programs in digital tv context (20)

Recommender helping viewers in their choice for educational programs in digit...
Recommender helping viewers in their choice for educational programs in digit...Recommender helping viewers in their choice for educational programs in digit...
Recommender helping viewers in their choice for educational programs in digit...
 
Using content-based filtering in a system of recommendation in the context of...
Using content-based filtering in a system of recommendation in the context of...Using content-based filtering in a system of recommendation in the context of...
Using content-based filtering in a system of recommendation in the context of...
 
BIPODITVR: brazilian interactive portable digital tv recommendation system
BIPODITVR: brazilian interactive portable digital tv recommendation systemBIPODITVR: brazilian interactive portable digital tv recommendation system
BIPODITVR: brazilian interactive portable digital tv recommendation system
 
T-learning for social inclusion
T-learning for social inclusionT-learning for social inclusion
T-learning for social inclusion
 
Application of recommendation techniques for brazilian portable interactive d...
Application of recommendation techniques for brazilian portable interactive d...Application of recommendation techniques for brazilian portable interactive d...
Application of recommendation techniques for brazilian portable interactive d...
 
BEACON: Can Digital Terrestrial Television lead to an inclusive society?
BEACON: Can Digital Terrestrial Television lead to an inclusive society?BEACON: Can Digital Terrestrial Television lead to an inclusive society?
BEACON: Can Digital Terrestrial Television lead to an inclusive society?
 
Iwssip application of recommendation techniques for brazilian portable inte...
Iwssip   application of recommendation techniques for brazilian portable inte...Iwssip   application of recommendation techniques for brazilian portable inte...
Iwssip application of recommendation techniques for brazilian portable inte...
 
Cable Television Video-On-Demand for Learner-Centered Instruction:on-manuscript
Cable Television Video-On-Demand for Learner-Centered Instruction:on-manuscriptCable Television Video-On-Demand for Learner-Centered Instruction:on-manuscript
Cable Television Video-On-Demand for Learner-Centered Instruction:on-manuscript
 
Business Models for Web TV - Research Report
Business Models for Web TV - Research ReportBusiness Models for Web TV - Research Report
Business Models for Web TV - Research Report
 
Applying actions model
Applying actions modelApplying actions model
Applying actions model
 
DTC meeting at the ITU, UN by Dr Danica Radovanovic
DTC meeting at the ITU, UN by Dr Danica RadovanovicDTC meeting at the ITU, UN by Dr Danica Radovanovic
DTC meeting at the ITU, UN by Dr Danica Radovanovic
 
MEGATRENDS SM DIPATANE.pdf
MEGATRENDS SM DIPATANE.pdfMEGATRENDS SM DIPATANE.pdf
MEGATRENDS SM DIPATANE.pdf
 
Web-TV an Innovative & Effective Digital Communication Tool for Think Tanks
Web-TV an Innovative & Effective Digital Communication Tool for Think TanksWeb-TV an Innovative & Effective Digital Communication Tool for Think Tanks
Web-TV an Innovative & Effective Digital Communication Tool for Think Tanks
 
CareNET Newsletter2
CareNET Newsletter2CareNET Newsletter2
CareNET Newsletter2
 
Zahid Hussain - Internet Tv Aug 2008 Poland
Zahid Hussain - Internet Tv Aug 2008 PolandZahid Hussain - Internet Tv Aug 2008 Poland
Zahid Hussain - Internet Tv Aug 2008 Poland
 
IAMCR 2004 - The Consequence of Digital TV for the Brazilian Media Professional
IAMCR 2004 - The Consequence of Digital TV for the Brazilian Media ProfessionalIAMCR 2004 - The Consequence of Digital TV for the Brazilian Media Professional
IAMCR 2004 - The Consequence of Digital TV for the Brazilian Media Professional
 
UNESCO-Pearson Initiative: Digital Inclusion Landscape Review
UNESCO-Pearson Initiative: Digital Inclusion Landscape ReviewUNESCO-Pearson Initiative: Digital Inclusion Landscape Review
UNESCO-Pearson Initiative: Digital Inclusion Landscape Review
 
Department of Health digital comms strategy 2011
Department of Health digital comms strategy 2011Department of Health digital comms strategy 2011
Department of Health digital comms strategy 2011
 
Innovative business model based on DVB-H and 3G synergy
Innovative business model based on DVB-H and 3G synergyInnovative business model based on DVB-H and 3G synergy
Innovative business model based on DVB-H and 3G synergy
 
AN ANALYSIS OF FACTORS AFFECTING THE ADOPTION OF DIGITAL TERRESTRIAL TELEVISI...
AN ANALYSIS OF FACTORS AFFECTING THE ADOPTION OF DIGITAL TERRESTRIAL TELEVISI...AN ANALYSIS OF FACTORS AFFECTING THE ADOPTION OF DIGITAL TERRESTRIAL TELEVISI...
AN ANALYSIS OF FACTORS AFFECTING THE ADOPTION OF DIGITAL TERRESTRIAL TELEVISI...
 

More from Elaine Cecília Gatto

A influência da Tecnologia em cada faixa etaria
A influência da Tecnologia em cada faixa etariaA influência da Tecnologia em cada faixa etaria
A influência da Tecnologia em cada faixa etariaElaine Cecília Gatto
 
Inteligência Artificial Aplicada à Medicina
Inteligência Artificial Aplicada à MedicinaInteligência Artificial Aplicada à Medicina
Inteligência Artificial Aplicada à MedicinaElaine Cecília Gatto
 
Além do Aprendizado Local e Global: Particionando o espaço de classes em prob...
Além do Aprendizado Local e Global: Particionando o espaço de classes em prob...Além do Aprendizado Local e Global: Particionando o espaço de classes em prob...
Além do Aprendizado Local e Global: Particionando o espaço de classes em prob...Elaine Cecília Gatto
 
Apresentação da minha tese de doutorado no EPPC
Apresentação da minha tese de doutorado no EPPCApresentação da minha tese de doutorado no EPPC
Apresentação da minha tese de doutorado no EPPCElaine Cecília Gatto
 
Como a pesquisa científica impacta o mundo real.pptx
Como a pesquisa científica impacta o mundo real.pptxComo a pesquisa científica impacta o mundo real.pptx
Como a pesquisa científica impacta o mundo real.pptxElaine Cecília Gatto
 
Explorando correlações entre rótulos para o particionamento do espaço de rótu...
Explorando correlações entre rótulos para o particionamento do espaço de rótu...Explorando correlações entre rótulos para o particionamento do espaço de rótu...
Explorando correlações entre rótulos para o particionamento do espaço de rótu...Elaine Cecília Gatto
 
Community Detection for Multi-Label Classification - Seminários UFSCar
Community Detection for Multi-Label Classification - Seminários UFSCarCommunity Detection for Multi-Label Classification - Seminários UFSCar
Community Detection for Multi-Label Classification - Seminários UFSCarElaine Cecília Gatto
 
Classificação Multirrótulo: Aprendizado de Correlações
Classificação Multirrótulo: Aprendizado de CorrelaçõesClassificação Multirrótulo: Aprendizado de Correlações
Classificação Multirrótulo: Aprendizado de CorrelaçõesElaine Cecília Gatto
 
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...Elaine Cecília Gatto
 
Community Detection Method for Multi-Label Classification
Community Detection Method for Multi-Label ClassificationCommunity Detection Method for Multi-Label Classification
Community Detection Method for Multi-Label ClassificationElaine Cecília Gatto
 
Mulheres na Campus Party assumir o feminismo ou não – Blogueiras Feministas.pdf
Mulheres na Campus Party assumir o feminismo ou não – Blogueiras Feministas.pdfMulheres na Campus Party assumir o feminismo ou não – Blogueiras Feministas.pdf
Mulheres na Campus Party assumir o feminismo ou não – Blogueiras Feministas.pdfElaine Cecília Gatto
 
Explorando Correlações entre Rótulos usando Métodos de Detecção de Comu...
Explorando Correlações entre Rótulos usando Métodos de Detecção de Comu...Explorando Correlações entre Rótulos usando Métodos de Detecção de Comu...
Explorando Correlações entre Rótulos usando Métodos de Detecção de Comu...Elaine Cecília Gatto
 
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...Elaine Cecília Gatto
 
Pipeline desdobramento escalonamento
Pipeline desdobramento escalonamentoPipeline desdobramento escalonamento
Pipeline desdobramento escalonamentoElaine Cecília Gatto
 
Resumo das Instruções de Desvio Incondicionais MIPS 32 bits
Resumo das Instruções de Desvio Incondicionais MIPS 32 bitsResumo das Instruções de Desvio Incondicionais MIPS 32 bits
Resumo das Instruções de Desvio Incondicionais MIPS 32 bitsElaine Cecília Gatto
 
Como descobrir e classificar coisas usando machine learning sem compilcação
Como descobrir e classificar coisas usando machine learning sem compilcaçãoComo descobrir e classificar coisas usando machine learning sem compilcação
Como descobrir e classificar coisas usando machine learning sem compilcaçãoElaine Cecília Gatto
 

More from Elaine Cecília Gatto (20)

A influência da Tecnologia em cada faixa etaria
A influência da Tecnologia em cada faixa etariaA influência da Tecnologia em cada faixa etaria
A influência da Tecnologia em cada faixa etaria
 
Inteligência Artificial Aplicada à Medicina
Inteligência Artificial Aplicada à MedicinaInteligência Artificial Aplicada à Medicina
Inteligência Artificial Aplicada à Medicina
 
Além do Aprendizado Local e Global: Particionando o espaço de classes em prob...
Além do Aprendizado Local e Global: Particionando o espaço de classes em prob...Além do Aprendizado Local e Global: Particionando o espaço de classes em prob...
Além do Aprendizado Local e Global: Particionando o espaço de classes em prob...
 
Apresentação da minha tese de doutorado no EPPC
Apresentação da minha tese de doutorado no EPPCApresentação da minha tese de doutorado no EPPC
Apresentação da minha tese de doutorado no EPPC
 
entrevista r7.pdf
entrevista r7.pdfentrevista r7.pdf
entrevista r7.pdf
 
Como a pesquisa científica impacta o mundo real.pptx
Como a pesquisa científica impacta o mundo real.pptxComo a pesquisa científica impacta o mundo real.pptx
Como a pesquisa científica impacta o mundo real.pptx
 
Empoderamento Feminino
Empoderamento FemininoEmpoderamento Feminino
Empoderamento Feminino
 
Explorando correlações entre rótulos para o particionamento do espaço de rótu...
Explorando correlações entre rótulos para o particionamento do espaço de rótu...Explorando correlações entre rótulos para o particionamento do espaço de rótu...
Explorando correlações entre rótulos para o particionamento do espaço de rótu...
 
Community Detection for Multi-Label Classification - Seminários UFSCar
Community Detection for Multi-Label Classification - Seminários UFSCarCommunity Detection for Multi-Label Classification - Seminários UFSCar
Community Detection for Multi-Label Classification - Seminários UFSCar
 
Classificação Multirrótulo: Aprendizado de Correlações
Classificação Multirrótulo: Aprendizado de CorrelaçõesClassificação Multirrótulo: Aprendizado de Correlações
Classificação Multirrótulo: Aprendizado de Correlações
 
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
 
Community Detection Method for Multi-Label Classification
Community Detection Method for Multi-Label ClassificationCommunity Detection Method for Multi-Label Classification
Community Detection Method for Multi-Label Classification
 
Mulheres na Campus Party assumir o feminismo ou não – Blogueiras Feministas.pdf
Mulheres na Campus Party assumir o feminismo ou não – Blogueiras Feministas.pdfMulheres na Campus Party assumir o feminismo ou não – Blogueiras Feministas.pdf
Mulheres na Campus Party assumir o feminismo ou não – Blogueiras Feministas.pdf
 
Curtinhas de sábado.pdf
Curtinhas de sábado.pdfCurtinhas de sábado.pdf
Curtinhas de sábado.pdf
 
Explorando Correlações entre Rótulos usando Métodos de Detecção de Comu...
Explorando Correlações entre Rótulos usando Métodos de Detecção de Comu...Explorando Correlações entre Rótulos usando Métodos de Detecção de Comu...
Explorando Correlações entre Rótulos usando Métodos de Detecção de Comu...
 
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
EXPLORANDO CORRELAÇÕES PARA O PARTICIONAMENTO DO ESPAÇO DE RÓTULOS EM PROBLEM...
 
Pipeline desdobramento escalonamento
Pipeline desdobramento escalonamentoPipeline desdobramento escalonamento
Pipeline desdobramento escalonamento
 
Cheat sheet Mips 32 bits
Cheat sheet Mips 32 bitsCheat sheet Mips 32 bits
Cheat sheet Mips 32 bits
 
Resumo das Instruções de Desvio Incondicionais MIPS 32 bits
Resumo das Instruções de Desvio Incondicionais MIPS 32 bitsResumo das Instruções de Desvio Incondicionais MIPS 32 bits
Resumo das Instruções de Desvio Incondicionais MIPS 32 bits
 
Como descobrir e classificar coisas usando machine learning sem compilcação
Como descobrir e classificar coisas usando machine learning sem compilcaçãoComo descobrir e classificar coisas usando machine learning sem compilcação
Como descobrir e classificar coisas usando machine learning sem compilcação
 

Recently uploaded

Kolkata Call Girl Bagbazar 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Bagbazar 👉 8250192130 ❣️💯 Available With Room 24×7Kolkata Call Girl Bagbazar 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Bagbazar 👉 8250192130 ❣️💯 Available With Room 24×7Riya Pathan
 
Air-Hostess Call Girls Diamond Harbour : 8250192130 High Profile Model Escort...
Air-Hostess Call Girls Diamond Harbour : 8250192130 High Profile Model Escort...Air-Hostess Call Girls Diamond Harbour : 8250192130 High Profile Model Escort...
Air-Hostess Call Girls Diamond Harbour : 8250192130 High Profile Model Escort...anamikaraghav4
 
College Call Girls New Alipore - For 7001035870 Cheap & Best with original Ph...
College Call Girls New Alipore - For 7001035870 Cheap & Best with original Ph...College Call Girls New Alipore - For 7001035870 Cheap & Best with original Ph...
College Call Girls New Alipore - For 7001035870 Cheap & Best with original Ph...anamikaraghav4
 
Cash Payment Contact:- 7028418221 Goa Call Girls Service North Goa Escorts
Cash Payment Contact:- 7028418221 Goa Call Girls Service North Goa EscortsCash Payment Contact:- 7028418221 Goa Call Girls Service North Goa Escorts
Cash Payment Contact:- 7028418221 Goa Call Girls Service North Goa EscortsApsara Of India
 
Call Girl Nagpur Roshni Call 7001035870 Meet With Nagpur Escorts
Call Girl Nagpur Roshni Call 7001035870 Meet With Nagpur EscortsCall Girl Nagpur Roshni Call 7001035870 Meet With Nagpur Escorts
Call Girl Nagpur Roshni Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur High Profile
 
Air-Hostess Call Girls Shobhabazar | 8250192130 At Low Cost Cash Payment Booking
Air-Hostess Call Girls Shobhabazar | 8250192130 At Low Cost Cash Payment BookingAir-Hostess Call Girls Shobhabazar | 8250192130 At Low Cost Cash Payment Booking
Air-Hostess Call Girls Shobhabazar | 8250192130 At Low Cost Cash Payment BookingRiya Pathan
 
Hot Call Girls In Goa 7028418221 Call Girls In Vagator Beach EsCoRtS
Hot Call Girls In Goa 7028418221 Call Girls In Vagator Beach EsCoRtSHot Call Girls In Goa 7028418221 Call Girls In Vagator Beach EsCoRtS
Hot Call Girls In Goa 7028418221 Call Girls In Vagator Beach EsCoRtSApsara Of India
 
Call Girls in Najafgarh Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Najafgarh Delhi 💯Call Us 🔝8264348440🔝Call Girls in Najafgarh Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Najafgarh Delhi 💯Call Us 🔝8264348440🔝soniya singh
 
Kolkata Call Girl Bara Bazar 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Bara Bazar 👉 8250192130 ❣️💯 Available With Room 24×7Kolkata Call Girl Bara Bazar 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Bara Bazar 👉 8250192130 ❣️💯 Available With Room 24×7Riya Pathan
 
Kolkata Call Girl Airport Kolkata 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Airport Kolkata 👉 8250192130 ❣️💯 Available With Room 24×7Kolkata Call Girl Airport Kolkata 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Airport Kolkata 👉 8250192130 ❣️💯 Available With Room 24×7Riya Pathan
 
VIP Call Girls Darjeeling Aaradhya 8250192130 Independent Escort Service Darj...
VIP Call Girls Darjeeling Aaradhya 8250192130 Independent Escort Service Darj...VIP Call Girls Darjeeling Aaradhya 8250192130 Independent Escort Service Darj...
VIP Call Girls Darjeeling Aaradhya 8250192130 Independent Escort Service Darj...Neha Kaur
 
Call Girl Nashik Amaira 7001305949 Independent Escort Service Nashik
Call Girl Nashik Amaira 7001305949 Independent Escort Service NashikCall Girl Nashik Amaira 7001305949 Independent Escort Service Nashik
Call Girl Nashik Amaira 7001305949 Independent Escort Service NashikCall Girls in Nagpur High Profile
 
VIP Call Girls Service Banjara Hills Hyderabad Call +91-8250192130
VIP Call Girls Service Banjara Hills Hyderabad Call +91-8250192130VIP Call Girls Service Banjara Hills Hyderabad Call +91-8250192130
VIP Call Girls Service Banjara Hills Hyderabad Call +91-8250192130Suhani Kapoor
 
Contact:- 8860008073 Call Girls in Karnal Escort Service Available at Afforda...
Contact:- 8860008073 Call Girls in Karnal Escort Service Available at Afforda...Contact:- 8860008073 Call Girls in Karnal Escort Service Available at Afforda...
Contact:- 8860008073 Call Girls in Karnal Escort Service Available at Afforda...Apsara Of India
 
Fun Call Girls In Goa 7028418221 Escort Service In Morjim Beach Call Girl
Fun Call Girls In Goa 7028418221 Escort Service In Morjim Beach Call GirlFun Call Girls In Goa 7028418221 Escort Service In Morjim Beach Call Girl
Fun Call Girls In Goa 7028418221 Escort Service In Morjim Beach Call GirlApsara Of India
 
Call Girls Service Bantala - Call 8250192130 Rs-3500 with A/C Room Cash on De...
Call Girls Service Bantala - Call 8250192130 Rs-3500 with A/C Room Cash on De...Call Girls Service Bantala - Call 8250192130 Rs-3500 with A/C Room Cash on De...
Call Girls Service Bantala - Call 8250192130 Rs-3500 with A/C Room Cash on De...anamikaraghav4
 
Kolkata Call Girls Service +918240919228 - Kolkatanightgirls.com
Kolkata Call Girls Service +918240919228 - Kolkatanightgirls.comKolkata Call Girls Service +918240919228 - Kolkatanightgirls.com
Kolkata Call Girls Service +918240919228 - Kolkatanightgirls.comKolkata Call Girls
 
VIP Call Girls Asansol Ananya 8250192130 Independent Escort Service Asansol
VIP Call Girls Asansol Ananya 8250192130 Independent Escort Service AsansolVIP Call Girls Asansol Ananya 8250192130 Independent Escort Service Asansol
VIP Call Girls Asansol Ananya 8250192130 Independent Escort Service AsansolRiya Pathan
 
Private Call Girls Bally - 8250192130 | 24x7 Service Available Near Me
Private Call Girls Bally - 8250192130 | 24x7 Service Available Near MePrivate Call Girls Bally - 8250192130 | 24x7 Service Available Near Me
Private Call Girls Bally - 8250192130 | 24x7 Service Available Near MeRiya Pathan
 

Recently uploaded (20)

Kolkata Call Girl Bagbazar 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Bagbazar 👉 8250192130 ❣️💯 Available With Room 24×7Kolkata Call Girl Bagbazar 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Bagbazar 👉 8250192130 ❣️💯 Available With Room 24×7
 
Air-Hostess Call Girls Diamond Harbour : 8250192130 High Profile Model Escort...
Air-Hostess Call Girls Diamond Harbour : 8250192130 High Profile Model Escort...Air-Hostess Call Girls Diamond Harbour : 8250192130 High Profile Model Escort...
Air-Hostess Call Girls Diamond Harbour : 8250192130 High Profile Model Escort...
 
College Call Girls New Alipore - For 7001035870 Cheap & Best with original Ph...
College Call Girls New Alipore - For 7001035870 Cheap & Best with original Ph...College Call Girls New Alipore - For 7001035870 Cheap & Best with original Ph...
College Call Girls New Alipore - For 7001035870 Cheap & Best with original Ph...
 
Cash Payment Contact:- 7028418221 Goa Call Girls Service North Goa Escorts
Cash Payment Contact:- 7028418221 Goa Call Girls Service North Goa EscortsCash Payment Contact:- 7028418221 Goa Call Girls Service North Goa Escorts
Cash Payment Contact:- 7028418221 Goa Call Girls Service North Goa Escorts
 
Call Girl Nagpur Roshni Call 7001035870 Meet With Nagpur Escorts
Call Girl Nagpur Roshni Call 7001035870 Meet With Nagpur EscortsCall Girl Nagpur Roshni Call 7001035870 Meet With Nagpur Escorts
Call Girl Nagpur Roshni Call 7001035870 Meet With Nagpur Escorts
 
Air-Hostess Call Girls Shobhabazar | 8250192130 At Low Cost Cash Payment Booking
Air-Hostess Call Girls Shobhabazar | 8250192130 At Low Cost Cash Payment BookingAir-Hostess Call Girls Shobhabazar | 8250192130 At Low Cost Cash Payment Booking
Air-Hostess Call Girls Shobhabazar | 8250192130 At Low Cost Cash Payment Booking
 
Hot Call Girls In Goa 7028418221 Call Girls In Vagator Beach EsCoRtS
Hot Call Girls In Goa 7028418221 Call Girls In Vagator Beach EsCoRtSHot Call Girls In Goa 7028418221 Call Girls In Vagator Beach EsCoRtS
Hot Call Girls In Goa 7028418221 Call Girls In Vagator Beach EsCoRtS
 
Call Girls in Najafgarh Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Najafgarh Delhi 💯Call Us 🔝8264348440🔝Call Girls in Najafgarh Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Najafgarh Delhi 💯Call Us 🔝8264348440🔝
 
Kolkata Call Girl Bara Bazar 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Bara Bazar 👉 8250192130 ❣️💯 Available With Room 24×7Kolkata Call Girl Bara Bazar 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Bara Bazar 👉 8250192130 ❣️💯 Available With Room 24×7
 
Kolkata Call Girl Airport Kolkata 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Airport Kolkata 👉 8250192130 ❣️💯 Available With Room 24×7Kolkata Call Girl Airport Kolkata 👉 8250192130 ❣️💯 Available With Room 24×7
Kolkata Call Girl Airport Kolkata 👉 8250192130 ❣️💯 Available With Room 24×7
 
Call Girls Chirag Delhi Delhi WhatsApp Number 9711199171
Call Girls Chirag Delhi Delhi WhatsApp Number 9711199171Call Girls Chirag Delhi Delhi WhatsApp Number 9711199171
Call Girls Chirag Delhi Delhi WhatsApp Number 9711199171
 
VIP Call Girls Darjeeling Aaradhya 8250192130 Independent Escort Service Darj...
VIP Call Girls Darjeeling Aaradhya 8250192130 Independent Escort Service Darj...VIP Call Girls Darjeeling Aaradhya 8250192130 Independent Escort Service Darj...
VIP Call Girls Darjeeling Aaradhya 8250192130 Independent Escort Service Darj...
 
Call Girl Nashik Amaira 7001305949 Independent Escort Service Nashik
Call Girl Nashik Amaira 7001305949 Independent Escort Service NashikCall Girl Nashik Amaira 7001305949 Independent Escort Service Nashik
Call Girl Nashik Amaira 7001305949 Independent Escort Service Nashik
 
VIP Call Girls Service Banjara Hills Hyderabad Call +91-8250192130
VIP Call Girls Service Banjara Hills Hyderabad Call +91-8250192130VIP Call Girls Service Banjara Hills Hyderabad Call +91-8250192130
VIP Call Girls Service Banjara Hills Hyderabad Call +91-8250192130
 
Contact:- 8860008073 Call Girls in Karnal Escort Service Available at Afforda...
Contact:- 8860008073 Call Girls in Karnal Escort Service Available at Afforda...Contact:- 8860008073 Call Girls in Karnal Escort Service Available at Afforda...
Contact:- 8860008073 Call Girls in Karnal Escort Service Available at Afforda...
 
Fun Call Girls In Goa 7028418221 Escort Service In Morjim Beach Call Girl
Fun Call Girls In Goa 7028418221 Escort Service In Morjim Beach Call GirlFun Call Girls In Goa 7028418221 Escort Service In Morjim Beach Call Girl
Fun Call Girls In Goa 7028418221 Escort Service In Morjim Beach Call Girl
 
Call Girls Service Bantala - Call 8250192130 Rs-3500 with A/C Room Cash on De...
Call Girls Service Bantala - Call 8250192130 Rs-3500 with A/C Room Cash on De...Call Girls Service Bantala - Call 8250192130 Rs-3500 with A/C Room Cash on De...
Call Girls Service Bantala - Call 8250192130 Rs-3500 with A/C Room Cash on De...
 
Kolkata Call Girls Service +918240919228 - Kolkatanightgirls.com
Kolkata Call Girls Service +918240919228 - Kolkatanightgirls.comKolkata Call Girls Service +918240919228 - Kolkatanightgirls.com
Kolkata Call Girls Service +918240919228 - Kolkatanightgirls.com
 
VIP Call Girls Asansol Ananya 8250192130 Independent Escort Service Asansol
VIP Call Girls Asansol Ananya 8250192130 Independent Escort Service AsansolVIP Call Girls Asansol Ananya 8250192130 Independent Escort Service Asansol
VIP Call Girls Asansol Ananya 8250192130 Independent Escort Service Asansol
 
Private Call Girls Bally - 8250192130 | 24x7 Service Available Near Me
Private Call Girls Bally - 8250192130 | 24x7 Service Available Near MePrivate Call Girls Bally - 8250192130 | 24x7 Service Available Near Me
Private Call Girls Bally - 8250192130 | 24x7 Service Available Near Me
 

Fie recommender helping viewers in their choice for educational programs in digital tv context

  • 1. Session S1E Recommender: Helping Viewers in their Choice for Educational Programs in Digital TV Context Paulo Muniz de Ávila, Elaine Cecília Gatto, Sergio Donizetti Zorzo pauloavila @pucpocos.com.br,elaine_gatto@dc.ufscar.br,zorzo@dc.ufscar.br Abstract - Currently in Brazil, a fundamental change is favorite program. In face of this situation, personalized taking place in TV: the migration from analogue to recommendation systems are necessary. digital TV system. This change has two main Different from EPG functions which allow basic search, implications: an increase in transmission capacity for a personalized TV system can create a profile for each TV new channels with the same bandwidth and the ability to viewer and recommend programs that best match this send applications with multiplexed audio-visual content. profile, avoiding the search in many EPG options to find the Brazilian government aims to exploit the transmission favorite program. Elementary and secondary education capacity for new channels offering programming created schools and universities generally seek to explore this new to distance learning and thereby promoting social model offering personalized content to their students. In this inclusion in the vast majority of the population. This context, a recommendation system is able to analyze the information overload demands mechanisms to help profile of a group of students, suggesting the educational students to browse and select what education programs content that best suits the needs of the group. are best suited to their current level. Personalized To make the benefits (new channels, interactive recommendation systems emerge as a solution to this applications) offered by the digital system possible, the TV problem, providing the viewer with educational viewers with analogical system need new equipment called programs relevant to his profile. In this paper we present set-top box (STB). STB is a device which works connected a personalized recommendation system, the to the TV and converts the digital sign received from the Recommender consistent with the reference provider to audio/video that the analogical TV can exhibit. implementation of the Brazilian digital TV system. To have the advantages offered by the digital TV, the STB Finally, we present the results obtained after using the needs a software layer which connects the hardware to the proposed system. interactive applications called middleware. The DTV Brazilian System middleware is Ginga [2,3]. It allows Key-words - Personalization, Multimedia, Recommendation declarative and procedural applications through its System, Digital TV, Middleware Ginga. components Ginga-NCL [2] and Ginga-J [3]. Ginga-NCL performs declarative application written in Nested Context INTRODUCTION Language (NCL) while Ginga-J can perform procedural Digital television has created new services, products, application based on JavaTM known as Xlets [4]. contents, channels and business models. The Brazilian This paper proposes an extension to Ginga middleware Digital TV System allows high quality audio and video, as through implementation of a new module incorporated to well as interactivity, creating different contents for users. Ginga Common Core called Recommender. The There are two main implications with Brazil Digital TV Recommender module is responsible for gathering, storing, System: the increase of the number of channels being processing and recommending TV education programs. To broadcasted with the same bandwidth and the possibility of develop the Recommender module, Ginga-NCL middleware sending multiplexed applications with the audio-visual developed by PUC-RIO (Pontifical Catholic University of content. As new channels emerge due to the transmission Rio de Janeiro) was used, implemented in C/C++ language increase, it is necessary to create ways that allow the TV with source code available under GPLv2 license and viewers to search among these channels. according with the patterns defined by the Brazilian system The Electronic Program Guide (EPG) helps the TV digital television [4]. viewers. However, as new channels are available, an TVDI IN BRAZIL AND EDUCATION information overload is unavoidable making the EPG system inappropriate. In Shangai [1], a big city in China, the TV One of the reasons to implement TVDi in the national operators provide different services (in the analogical territory is its potential to social inclusion. In Brazil, in many system, channels), and this number has been increasing at a cases, the open TV is the only source of information for 20% rate per year. Thus, the traditional EPG system became people who do not frequently read newspaper, magazine or unattractive because it takes too long for the viewers to any other kind of printed media. If we consider that the search among hundreds of options available to find their access to written information is low and that the information 978-1-4244-6262-9/10/$26.00 ©2010 IEEE October 27 - 30, 2010, Washington, DC 40th ASEE/IEEE Frontiers in Education Conference S1E-1
  • 2. Session S1E transmitted through TV newscasts is the biggest link features anymore due particularly to the financial survival of between the world and the daily routine of Brazilian people, theses broadcastings. It is possible to note, according to we have many reasons not to ignore the reach power of this legislation, that the programming only admits transmission technology. If it is correctly and consciously explored, with of programs with educative-cultural purposes. However, the help of interactive resources, TVDi can represent a there is the option to recreational, informative or sport powerful tool to have access to differentiated educational programs considered educative-cultural since they present knowledge at the same time it can include Brazilian citizens instructive elements or educative-cultural focus identified in digitally excluded nowadays. Thus, it can be said that in its presentation. Brazil, the access to the Internet is low and high-class people Digital TV implantation in Brazil has been advancing. Some are those who have more access to it and participate obstacles – among them the situation of commercial somehow in the educational scenario. The low number of broadcastings, political interests, influences (and models) of personal computers and the high number of TV sets in digital television international systems, legislation ruling the Brazilian houses defend the efforts to use all TVDi potential radio broadcasting – still prevent its complete operation, but in issues in the educational extent. If public policies are well when it is defined, a social participation never seen before in structured, TVDi can reinforce a new educational paradigm, other historical moments can take place in Brazil, ensuring allowing the entire population to have access to Internet access to information and culture. [6] resources, video, images, sounds, interactivity to introduce new knowledge, entertainment, education, leisure, services. RELATED WORKS It can allow the unlimited access to written and audiovisual There are several recommendation systems for DTV (Digital information. As the great part of Brazilian population has a Television) designed to offer a distinct personalization limited access to information and Internet, and considering service and to help TV viewers to deal with the great the fact that the TV is the durable good which is in almost all quantity of TV programs. Some systems related to the Brazilian houses, we can consider the TVDi a way to current work are presented here. significantly change the perspective of Brazilian distance The AIMED system proposed by [7], presents a learning. Even knowing that the TVDi inclusion in Brazil recommendation mechanism that considers some TV viewer will not solve the social inclusion problem, it is certain that characteristics as activities, interests, mood, TV use all its power can improve the digital inclusion, for it will background and demographic information. These data are ensure the information access, services and education to inserted in a neural network model that infers the viewers’ people with low purchasing power. [5] preferences about the programs. Unlike the work proposed EDUCATIVE BROADCASTING IN BRAZIL in this paper, which uses the implicit data collection, in the AIMED system, the data are collected and the system is set According to the Communication Department, educative trough questionnaires. This approach is doubtful, mainly broadcasting is the Sound Broadcasting Service (radio) or when limitations imposed to data input in a DTV system are Sounds and Images Services (TV) intended for the considered. transmission of educative-cultural programs which, besides In [8] a method to discover models of multiuser environment performing together with teaching systems of any level or in intelligent houses based on users’ implicit interactions is modality, aims the basic and higher education, the presented. This method stores information in logs. So, the permanent education and the professional education, besides logs can be used by a recommendation system in order to comprehending educational, cultural, pedagogical and decrease effort and adapt the content for each TV viewer as professional orientation activities. The execution of well as for multiuser situations. Evaluating the TV viewers’ broadcasting services with exclusively educative purposes is background of 20 families, it was possible to see that the granted to legal entities with internal public right, including accuracy of the proposed model was similar to an explicit universities, which will be given the preference to obtain the system. This shows that collecting the data in an implicit grant, and foundations privately established and others way is as efficient as the explicit approach. In this system, Brazilian universities. the user has to identify himself in an explicit way, using the The first educative broadcasting station, the University TV remote control. Unlike this system, the proposal in this paper of Pernambuco pertaining to the Education Department, was aims at promoting services to the recommendation systems on TV in 1967. Until 1980´s, educative TV broadcasting in for a totally implicit multiuser environment. Brazil gave priority to essentially educative programs and in In [9], a program recommendation strategy for multiple TV 1997, the Brazilian Association of Public, Educative and viewers is proposed based on the combination of the Cultural Broadcasting (ABEPEC) was created. In 1999, the viewer’s profile. The research analyzed three strategies to participant broadcastings created the RPTV (Public TV perform the content recommendation and provided the Network) which aims at establishing a common and choice of the strategy based on the profile combination. The mandatory programming guide to the associated results proved that the TV viewers’ profile combination can broadcastings. Today, the programming is different from reflect properly in the preferences of the majority of the that one in the beginning of educative broadcasting members in a group. The proposal in this paper uses an transmissions, that is, it does not have the strict educative approach similar to a multiuser environment, however, 978-1-4244-6262-9/10/$26.00 ©2010 IEEE October 27 - 30, 2010, Washington, DC 40th ASEE/IEEE Frontiers in Education Conference S1E-2
  • 3. Session S1E besides the profile combination, the time and day of the a set of modules responsible for the data processing, week are also considered. information filtering in the transport stream. It is the In [1] a personalized TV system is proposed loaded in the architecture core; Stack protocol layer responsible for STB compatible with the Multimidia Home Plataform supporting many communication protocols like HTTP, RTP (MHP) model of the digital television European pattern. and TS. According to the authors, the system was implemented in a commercial solution of the MHP middleware, and for that, implemented alterations and inclusions of new modules in this middleware. Offering recommendation in this system requires two important information that must be available: programs description and the viewer visualization behavior. The description of the programs is obtained by demultiplexing and decoding the information in the EIT (Event Information Table) table. EIT is the table used to transport specific information about programs, such as: start time, duration and description of programs in digital television environments. The viewing behavior is collected monitoring the user action with the STB and the later persistence of this information in the STB. The work of [1] is similar to the work proposed in this paper. The implicit collection of data, along with the inclusion of a new module FIGURE 1 – GINGA MIDDLEWARE ARCHITECTURE (ADJUSTED in the middleware architecture, is an example of this WITH THE RECOMMENDATION SYSTEM) similarity. In [10], the Personalized Electronic Program Guide is The proposed system extends the Ginga middleware considered a possible solution to the information overload functionalities including new services in the Ginga Common problem, mentioned in the beginning of this work. The Core layer. The Recommender module is the main part of authors compared the use of explicit and implicit profile and the recommendation system and it is inserted in the proved that the indicators of implicit interests are similar to Common Core layer of Ginga-NCL architecture. The the indicators of explicit interests. The approach to find out Recommender module is divided in two parts. The first one the user’s profile in an implicit way is adopted in this work describes the components integrated to the source code of and it is about an efficient mechanism in the context of the middleware such as Local Agent, Schedule Agent, Filter television environment, where the information input is Agent and Data Agent. The second part describes the new performed through remote control, a device that was not component added to the STB: Sqlite [13], a C library which designed to this purpose. implements an attached relational database. Figure 2 In [11], the AVATAR recommendation system is presented, presents the Recommender module architecture. compatible to the European MHP middleware. The authors I. Implemented Modules propose a new approach, where the recommendation system is distributed by broadcast service providers, as well as an This subsection describes the modules added to the Ginga- interactive application. According to the authors, this NCL middleware source code and the extensions approach allows the user to choose among different implemented to provide a better connection between recommendation systems, what is not possible when we middleware and the recommendation system. have an STB with a recommendation system installed in Local Agent is the module responsible for constant plant. The AVATAR system uses the approach of implicit monitoring of the remote control. Any interaction between collection of user profile and proposes modifications in the the viewer and the control is detected and stored in the MHP middleware to include the monitoring method. The database. The Local Agent is essential for the Naïve Bayes [12] is used as a classification algorithm and recommendation system that uses implicit approach to one of the main reasons for that is the low use of STB perform the profile. resources. Scheduler Agent is the module responsible for periodically SYSTEM OVERVIEW request the data mining. Data mining is a process that demands time and processing, making its execution The recommendation system proposed in this paper is based impracticable every time the viewer requests a on Ginga middleware. As mentioned before, the version recommendation. Scheduler Agent module guarantees a new used was the open source version of Ginga-NCL processing every 24 hours preferably at night, when the STB middleware. Figure 1 presents its architecture consisting of is in standby. three layers: Resident applications responsible for the exhibition (frequently called presentation layer); Ginga Common Core, 978-1-4244-6262-9/10/$26.00 ©2010 IEEE October 27 - 30, 2010, Washington, DC 40th ASEE/IEEE Frontiers in Education Conference S1E-3
  • 4. Session S1E items. For example, the system can be used to create a top- 10 question topic; the students would classify extra material with a grade and the best extra materials would be recommended. It would be also possible to have a top-10 favorite and a top-10 best students. Moreover, the system could also provide a way to look for old content interesting for the user to improve what is being studied at that moment. METHODOLOGY AND TESTS User history and EPG data are necessary to perform the tests. These data were provided by IBOPE (Brazilian Institute of Public Opinion and Statistics) [14] through a treatment process almost entirely manual in order to be in accordance to the standard format which must be used in the Brazilian digital TV system and also in the tests. Many technologies have been arising with the aim at identifying behavior standards and its application in the FIGURE 2 – RECOMMENDER MODULE ARCHITECTURE personalization. The recommendation systems operation is found on these techniques and the most used are the Collaborative Filtering and Content-Based Filtering which Mining Agent is the module that accesses the information in includes several algorithms for each one. A the viewer’s behavior background and the programming data recommendation system can use only one technique or two from the EIT and SDT tables stored in cache to perform the together, becoming a hybrid system. data mining. In order to process the data mining, the Mining In order to study, analyze and choose an algorithm to be module has direct access to the database and recovers the used in Technical module, some information filtering TV viewer’s behavior background. From the point of view algorithms were tested. The tests were performed in three of the system performance, this communication between steps. In the first step, tests were performed with Apriori mining module and user database is important. Without this algorithm. In the second step, the forecast method was used, communication, it would be necessary to implement a new applying Cosine as measure of similarity. The third step was module responsible for recover the database information and to compare the results and the operation with both then make such data available to the mining algorithm. The algorithms, analyzing the facilities and difficulties, second data set necessary to make possible the data mining especially for the implementation. is the program guide. The program guide is composed by The association techniques algorithms identify information sent by providers through EIT and SDT tables. associations between the data registers which are related in These tables are stored in cache and are available to be some way. The basic premise finds elements which imply recovered and processed by the Mining module. Ginga-NCL the presence of others in a same operation aiming at Middleware does not implement storage mechanism in cache determining which are related. The association rules of EIT and SDT tables. This functionality was implemented interconnect objects trying to show characteristics and by the Recommender system. tendencies. The association discoveries present trivial and non trivial association. The data was adapted in order to be Filter Agent & Data Agent The raw data returned by the used in Apriori algorithm, that is, it was submitted to a pre- Mining Agent module need to be filtered and later stored in processing phase. The user history was created from IBOPE the viewer’s database. The Filter Agent and Data Agent data. For the implementation, it is not necessary that the data modules are responsible for this function. The Filter Agent go through adjustments, as it will be collected in the correct module receives the data from the mining provided by the format to be used. The results were satisfactory verifying Mining Agent and eliminates any information that is not that Apriori can be applied to the system for it can be important keeping only those which are relevant to the adapted to the system needs. [15, 16] recommendation system such as the name of the program, The Cosine is a similarity measure, a forecast method time, date, service provider and the name of the service. The which calculates the similarity between items and users, Data Agent module receives the recommendations and stores consults similar items to a given item and matches item them in the viewer’s database. content and user profile. The data also had to be adjusted to be used with Cosine. Database in sqlite was used with the If there were many educative programs on open TV, it EPG and the user history. From these two tables, it was would be very useful to recommend other educative possible to derive two more, one with the profile of the programs. However, “educative” is one of the many TV program watched by the user and other with the profile of program categories. The system can be used inside a genres. It was necessary that the EPG passed through a distance learning system to recommend several types of modification which should also occur in the implementation. 978-1-4244-6262-9/10/$26.00 ©2010 IEEE October 27 - 30, 2010, Washington, DC 40th ASEE/IEEE Frontiers in Education Conference S1E-4
  • 5. Session S1E A new table was created, identical to the EPG table, but added with fields containing the genres names. According to the adjustment of the program in the genres, these fields were populated with 0 or 1, becoming a matrix. From these tables it was possible to find the Cosine for the programs and genres, the profile and what could be recommended to the user. The results from the Cosine were also satisfactory confirming that this technique can be applied to the system for it can be adjusted to the system needs. [17, 18] ANALYSIS During the tests, it was possible to note some particularities. Our system recommends contents based on the programs genres and our analyses were performed according to this standard. With Apriori algorithm, the data are collected in the correct format to be used. For the Cosine, FIGURE 3. ACCURACY OF THE RECOMMENDATION SYSTEM the EPG needs to be changed to a matrix before starting the process of discovering profiles and recommendations. Figure 3 presents the results obtained after 4 weeks of In a desktop, the feedback of the Cosine calculation is monitoring considering the best value obtained among the 8 faster in relation to the feedback of Apriori association rules. schools analyzed. It is clear that on the first weeks, as the However, further studies about these algorithms processing collected data were few, Apriori algorithm did not extract in these devices are still being performed. Apriori is able to relevant information from the preferences of the group. With discover the profile from the standards, but to select the the data increase in the visualization background on the third programs to be recommended, another technique must be and fourth week, the algorithm obtained better results and used and the Cosine can find both the profile and the the index of recommendation acceptance increased. recommendations. The Cosine cannot discover these characteristics, but reaches our goal. In order to discover behaviors similar to the association rules, it is necessary to consult the databank. Apriori output must be operated in order to give the correct user profile, that is, the rules must be understood, and that is very hard concerning implementation. The Cosine output is clearer; the result straightly reaches intended goal, allowing the output to be used without the need of a post-treatment. Regarding the input, there is no need of treatment for Apriori, since all data will be used as they are collected. However, for the Cosine, whenever the EPG is updated, the table containing the EPG matrix must be changed according to the new EPG, becoming something hard to work. The profile of the genres founded by both algorithms is similar. RESULTS In order to measure the evolution of the recommendation FIGURE 4 ACCURACY OF THE RECOMMENDATION SYSTEM PER SCHOOL offered to the students viewer, the following formula was applied: Figure 4 presents the accuracy per school. The main Ef = (α / β) 100 (1)  characteristic of the schools is the socioeconomics difference among them. The conclusion is that Apriori algorithm had a Where Ef is the efficacy of the recommendation system, good performance unrestricted to the students’ ranging from 0 to 100, α is the recommendation number ´socioeconomic profile. accepted by the students viewers and β is the number of recommendation presented. In order to monitor these data provided by IBOPE were used. The validation adopted an accuracy formula presented in (1). 978-1-4244-6262-9/10/$26.00 ©2010 IEEE October 27 - 30, 2010, Washington, DC 40th ASEE/IEEE Frontiers in Education Conference S1E-5
  • 6. Session S1E [3] Souza Filho, G. L., Leite, L. E. C., Batista, C. E. C. F. 2007, “Ginga-J: The Procedural Middleware for the Brazilian Digital TV System.”, Journal of the Brazilian Computer Society, v. 12, n. 4, p. 47-56, March 2007. [4] Ginga-NCL Virtual STB, (March 2009), Available at: http://www.ncl.org.br/ferramentas/index_30.html. [5] Silva, Dirceu et al. Possibilidades educativas e de inclusão social e digital com a TVDi: uma breve análise do cenário brasileiro. Universidade Estadual de Campinas, Brasil. Available in: http://www.google.com.br/url?sa=t&source=web&ct=res&cd=1&ved =0CAYQFjAA&url=http%3A%2F%2Fwww.rieoei.org%2Fdeloslecto res%2F2907Veraszto.pdf&ei=4ar5S- eeM8imuAeA1PG9Dg&usg=AFQjCNHaebMawhyd- xbrBsw0JjSHbTr7LQ&sig2=VGSGkTroJZu8t3NYiHA9_A. Acess in 2010-05-20. [6] Fort, Mônica Cristine. Televisão + Educação = Televisão Educativa. Available in: http://www.google.com.br/url?sa=t&source=web&ct=res&cd=4&ved =0CBUQFjAD&url=http%3A%2F%2Fencipecom.metodista.br%2Fm ediawiki%2Fimages%2Fb%2Fbf%2FGT10_- _008.pdf&ei=N8n5S5TaDYmHuAfp5Py9Dg&usg=AFQjCNFLgg9ng 2elo7UJAjo9dpf8-3I4hg&sig2=RrpI9k-iotN31hCBrc6yNw. Acess in FIGURE 5 – RECOMMENDERTV SYSTEM 2010-05-20. [7] S. H. Hsu, M. H. Wen, H. C. Lin, C. C. Lee, C. H. Lee. 2007. Figure 5 shows Recommender system. The application “AIMED – A personalized TV Recommendation System” in Proc 2007 Interactive TV: A Shared Experience. 5th European Conference, used as front-end is written in NCL and allows the students EuroITV 2007, Amsterdam, the Netherlands. to search the recommendation list selecting the education [8] Vildjiounaite, E., Kyllonen, V., Hannula, T. and Alahuhta, P. 2008. program. Unobtrusive Dynamic Modelling of TV Program Preferences. In Proceedings of the Changing Television Environments, 6th European CONCLUSION Conference, EuroITV 2008, pages 82-91. [9] Zhiwen, Y., Xingshe, Z., Yanbin, H. and Jianhua, G. 2006. TV With the appearance of digital TV, a variety of new program recommendation for multiple viewers based on user profile services (in the analogical system, channels) will be merging. In Proceedings of the User Modeling and User-Adapted available. This information overload requires the Interaction, pages 63-82. Publishing Springer Netherlands. implementation of new mechanisms to offer facilities to the [10] O’Sullivan, D., Smyth, B., Wilson, D. C., McDonald, K. and Smeaton, A. 2004. Interactive Television Personalization: From students looking for their education programs. These new Guides to Programs. Personalized Digital Television: Targeting mechanisms suggesting the viewers programs are known as Programs to Individual Viewers. L. Ardissono, A. .Kobsa and M. recommendation systems. A recommendation system Maybury editors, pages 73-91, Kluwer Academic Publishers compatible with Ginga middleware is presented in this paper [11] Blanco-Fernandez, Y., Pazos-Arias, J., Gil-Solla, A., Ramos-Cabrer, and it is implemented according to the standards of the M.,Lopes-Nores, M., Barragans-Martinez, B. 2005. AVATAR: a Multi-agent TV Recommender System Using digital television Brazilian system. The recommendation MHP Applications. In: IEEE International Conference on E- system was modeled considering the current characteristics Technology, E-Commerce and E-Service (EEE '05), pp. 660-665. of the television, and this model can be adjusted to other [12] Wu, X., Kumar, V., Ross Quinlan, J., Ghosh, J., Yang, Q., Motoda, standards and also to new portable devices which will be on H., McLachlan, G. J., Ng, A., Liu, B., Yu, P. S., Zhou, Z., Steinbach, M., Hand, D. J., and Steinberg, D. 2007. Top 10 algorithms in data the market. At last, future works can include algorithms of mining. Knowl. Inf. Syst. 14, 1 (Dec. 2007), 1-37. DOI= collaborative filtering and also a new architecture using http://dx.doi.org/10.1007/s10115-007-0114-2 client-server, providing and offering other kinds of [13] Sqlite, (March 2010) Available at: http://www.sqlite.org/. personalization services for the users. [14] “IBOPE”. Available in: http://www.ibope.com.br. Access in December 2009. ACKNOWLEDGMENT [15] Witten, I. H.; Frank, E. Data Mining: Practical Machine Learning Tools and Techniques, 2nd Edition, Morgan Kaufmann, 525 pages, We thank IBOPE for providing real data about the June 2005. electronic program guide and also the viewer’s behavior data [16] Gatto, Elaine C.; Zorzo, Sergio D. “Sistema de Recomendação para from March, 05, 2009 to March, 19, 2009. TVDPI,” in 8th International Information and Telecommunication Technologies Symposium. Florianópolis, Santa Catarina, Brasil. 09- REFERENCES 11/12/2009. [17] Torres, Roberto. “Personalização na Internet.” Novatec Editora. 2004. [1] H. Zhang, S. Zheng and J. Yuan. 2005, “A Personalized TV Guide 158p. System Compliant with MHP”, IEEE Transactions on Consumer [18] Gatto, Elaine C.; Zorzo, Sergio D. Application of recommendation Electronics, pages 731-737, Vol. 51, No. 2, MAY 2005. techniques for Brazilian Portable Interactive Digital TV. In: IWSSIP [2] L.F.G. Soares, R.F. Rodrigues, M.F. Moreno. 2007, “Ginga-NCL: The 2010 - 17th International Conference on Systems, Signals and Image declarative Environment of the Brazilian Digital TV System”, Journal Processing. June 17-19, 2010, Rio de Janeiro, Brazil. of the Brazilian Computer Society. V.12, n.4, p.37-46, March 2007. 978-1-4244-6262-9/10/$26.00 ©2010 IEEE October 27 - 30, 2010, Washington, DC 40th ASEE/IEEE Frontiers in Education Conference S1E-6