SlideShare a Scribd company logo
1 of 26
Hooman Homayounfard, Paul Kennedy & Robin Braun
Faculty of Engineering & Information Technology
University of Technology, Sydney
{hooman.homayounfard, paul.kennedy, Robin.Braun}@uts.edu.au
NARGES: Loss prediction model based on
symbolic delay forecasting
27 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Agenda
 Introduction
 Background
 NARGES Model
 Model Evaluation
 Results
 Conclusions
227 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Introduction: Internet
FISH Problem
327 September 2013
Can we risk by sending valuable data stream to a
highly congested “Hi-Performance” link?
AS 2AS 1 Internet Cloud
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Introduction
 Quality of Service (QoS) time-series
associated to a link within the Internet
 The “time cost” of numerical time-series
analysis methods for “trivial forecasting
accuracy” is prohibitive.
 Researchers suggest a transition from
computing with numbers to the manipulation
of perceptions in the form of linguistic
variables (Zadeh, 2001; Keogh et al., 2005; Batyrshin &
Sheremetov, 2008).
427 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Introduction: Quality of
Service (QoS)
 Related network parameters that have impact on
the QoS of the audio/video applications
Delay
Jitter
Packet-loss
 There is correlation between delay (and jitter)
values and the probability of packet-loss
occurrence in a link. (Roychoudhuri 2007)
527 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
 “Computing with words as a transformation of an initial data
set (IDS) into a terminal dataset (TDS)” (Zadeh ,1999 )
 Computing with Words (CW): p and q are prepositions
stated in human linguistic terms
Background (I)
6
CWIDS TDS
Initial data set: (p) Terminal data set: (q)
27 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Background (II)
7
 Let X, Y be independent random variables
taking values in a finite set of linguistic
terms V = {v1,v2,v3,…,vn} with probabilities
p1,…, pn and q1,…, qn, respectively.
27 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
NARGES Model Overview
827 September 2013
Forecasted
Delay
Forecasted
Jitter
Forecasted
Delay
Forecasted
Delay & Jitters
(t)
Packet-Loss Time Series Data Network (used for training MLP ANN)
Measured
QoS Time Series
(t-1 & t-2)
Forecasting
Module
(HDAX)
Pattern
Classification &
Approximation
Pattern Database/Look-up Table
of triplet i-j-k Trend Patterns
and their observed frequencies
Pattern
Frequencies
Retrieval for
Approximation
Storing
Patterns
& Frequencies
PDB Predicted
Packet-Loss
(t+1)
Delayed
Incoming
QoS Data
(Delay, Jitter,
Packet-loss)
MLP ANN
(Predictive
Module)
Predicted
Packet-Loss
Forecasted
Jitter
Measured
Jitter
Measured
Delay
 NARGES cosist of two subcomponent:
 Forecasting Module (HDAX)
 Predictive Module (MLP)
Forecasting Module (HDAX)
Forecasted
Delay
Forecasted
Delay & Jitters
(t)
Measured
QoS Time Series
(t-1 & t-2)
Forecasting
Module
(HDAX)
Pattern
Classification &
Approximation
Pattern Database/Look-up Table
of triplet i-j-k Trend Patterns
and their observed frequencies
Pattern
Frequencies
Retrieval for
Approximation
Storing
Patterns
& Frequencies
PDB
Delayed
Incoming
QoS Data
(Delay, Jitter,
Packet-loss)
Forecasted
Jitter
Measured
Jitter
Measured
Delay
927 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
HDAX: Deterministic
Mapping Function (DMF)
Basic Patterns are described in terms of
linguistic variables by Deterministic mapping
Function (DMF) like
Alphabet = {SI, I, P, D, SD, OUT}
 DMF: The scale of time-series trends
27 September 2013 10
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
HDAX: Basic & Composite
Patterns
 Pattern Definition
a. Basic Pattern
b. Composite Pattern
27 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
 Transition from Number to Linguistic
Values
HDAX: Transition from
Number to Linguistic Values
12
SI
SI SD
SDSI
t
Y
Time
Values
f : a numerical (crisp) function
f *: approximation of
f
If Time is t then Y is Yt
Yt is a Linguistic value assigned to a Y numerical value at time t
D D P P
SI
27 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
6000 -
-
5000 -
-
4000 -
-
3000 -
-
2000 -
-
1000 -
-
Y-Axis
Time
SD(4)
I (1) D (3) P (0)
I (1)
P (0) SI (2) I (1)
SI (2)
SD (4)
SI (2)
I (1) D (3)
Time-SeriesValues
End-to-endDelay/Jitter
SD (4)
I – D – P
P – SI – I
Pattern Database
(Runtime: Lookup Table)
.
.
.
.
.
.
.
.
.
P – P – P
SD – SD – SD
.
.
.
Y-Axis
yt-2
yt-1
yt
yt-3
HDAX: Training & Simulation
13
Previous(i)
Current(j)
Next(k)
• Training time: The i − j − k patterns are counted to populate the
corresponding i − j − k frequency entries in the pattern database.
• Simulation time:
•The frequencies stored in Pattern Database (Look-up Table) is used for finding
the most frequent pattern having previous two time-series trends.
• This estimated trend is used to approximate the yt values using Euler method.
27 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Predictive Module (MLP)
1427 September 2013
Forecasted
Jitter
Forecasted
Delay
Forecasted
Delay & Jitters
(t)
Predicted
Packet-Loss
(t+1)
MLP ANN
(Predictive
Module)
Predicted
Packet-Loss
Measured
QoS Time Series
(t-1 & t-2)
Delayed
Incoming
QoS Data
(Delay, Jitter,
Packet-loss)
Measured Packet-Loss Time Series
(used for training MLP ANN)
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
 Parameter settings is done based on the
optimum MLP results of
 Loss Function (NRMSE)
 Performance (speed)
 MLP Training
The Levenberg-Marquardt optimisation process is
based upon Bayesian regularization that
minimises:
 MSE, and
 weights,
MLP: Parameter Setting
& Training
1527 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Data Sources (I)
 Datasets:
 Archives are provided by Napoli University
“Federico II” generated by D-ITG Traffic Generator
http://www.grid.unina.it/Traffic/Traces/dtraces
(36 Datasets)
 Network logs simulated on OPNET Modeller
http://www.opnet.com/solutions/network_rd/modeler
.html
(10 Datasets)
 Data used for:
 Training (25%)
 Simulation (75%).
1627 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Model Evaluation
 Evaluation Benchmark: Autoregressive Moving
Average (ARMA)
 Measurements
 Error Measurement (Loss Function):
Normalised Root Mean Square Error
(NRMSE)
Speed Measurement
 Cross-Correlation Function (CCF)
 Non-parametric Holm’s test used to interpret the
results. We Tested the similarity of algorithms for
accuracy, speed and cross-correlation (CCF) over Delay,
Jitter and Packet-Loss
1727 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Results: Experiments
D-ITG Datasets
 Holm Table for α = 0.05. Models printed in bold are
statistically significantly better.
1827 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Results: Experiments
OPNET Datasets
 Holm Table for = 0.05. Models printed in bold are
statistically significantly better.
1927 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Conclusions & Future Work
 Conclusions
 NARGES results were statistically shown to have
significant correlation and precision compared to the
measured values on a link.
 The results demonstrated the quality and precision
improvement of NARGES compared to ARMA model.
 Future Work
 Implementing the model within a co-simulation
environment for online model evaluation.
 Info. on: www-staff.it.uts.edu.au/~hoomanhm
2027 September 2013
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
Thanks for your attention
Hooman Homayounfard
Hooman.Homayounfard@uts.edu.au
27 September 2013 21
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
References
 Batyrshin, I. Z. & Sheremetov, L. B. (2008), `Perception-based approach
to time series data mining', Applied Soft Computing Journal 8(3),1211-
1221.
 Biaz, S. & Vaidya, N. H. (1998), Distinguishing congestion losses from
wireless transmission losses: A negative result, in `7th Int. Conf.
Computer Communications and Networks', Lafayette, LA, pp. 722-731.
Proceedings of the Seventh International Conference on Computer
Communications and Networks (IC3N).
 Botta, A., Dainotti, A. & Pescap, A. (2007), Multi-protocol and multi-
platform traffic generation and measurement, in `IEEE INFOCOM 2007',
Anchor-age, Alaska, USA.
 Box, G. E. P., Jenkins, G. M. & Reinsel, G. C. (2008), Time series
analysis: forecasting and control, Wiley series in probability and
statistics, 4th edn, John Wiley, Hoboken, N.J.
 Brakmo, L. S., O'Malley, S. W. & Peterson, L. L. (1994), `TCP Vegas:
New techniques for congestion detection and avoidance', ACM
SIGCOMM Computer Communication Review 24(4), 24-35.
 Bui, V., Zhu, W., Pescape, A. & Botta, A. (2007), Long horizon end-to-
end delay forecasts: A multi-step-ahead hybrid approach, in `12th IEEE
Symposium on Computers and Communications, 2007. ISCC 2007',
IEEE, pp. 825-832.
27 September 2013 22
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
References
 Carter, R. L. & Crovella, M. E. (1996), Measuring bottleneck link speed
in packet-switched networks, Vol. 27, 28 of Performance evaluation,
Boston University, Boston, MA, USA.
 Chatfield, C. (2001), Time-series forecasting, Chap-man & Hall.
 Debenham, J. & Simoff, S. (2007), Informed agents: Integrating data
mining and agency, in C. Boukis, L. Pnevmatikakis & L. Polymenakos,
eds, `International Federation for Information Processing-IFIP', Vol. 247,
Springer-Verlag, Boston, pp. 165-173.
 Dovrolis, C., Ramanathan, P. & Moore, D. (2001), What do packet
dispersion techniques measure?, in `INFOCOM 2001. Twentieth Annual
Joint Conference of the IEEE Computer and Communications Societies.
Proceedings. IEEE', Vol. 2, pp. 905-914.
 Handley, M. (1997), An examination of MBone performance, Technical
report, ISI Research Report ISI/RR-97.
 Hannan, E. & Rissanen, J. (1982), `Recursive estimation of mixed
autoregressive-moving average order', Biometrika 69(1), 81-94.
27 September 2013 23
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
References
 Hatonen, K., Klemettinen, M., Mannila, H., Ronkainen, P. & Toivonen, H.
(1996), TASA: Telecommunication alarm sequence analyzer or how to
enjoy faults in your network, in `Network Operations and Management
Symposium, 1996., IEEE', Vol. 2, pp. 520-529.
 Hermanns, O. & Schuba, M. (1996), `Performance investigations of the
IP multicast architecture', Computer Networks and ISDN Systems 28(4),
429-439.
 Jain, M. & Dovrolis, C. (2002), End-to-end available bandwidth:
Measurement methodology, dynamics, and relation with TCP
throughput, in `Proceedings of the 2002 SIGCOMM conference 4', Vol.
32, ACM New York, NY, USA, pp. 295-308.
 Jain, R. (1989), `A delay-based approach for congestion avoidance in
interconnected heterogeneous computer networks', SIGCOMM Comput.
Commun. Rev. 19(5), 56-71. 74686.
 Jiang, W. & Schulzrinne, H. (2000), `Modeling of packet loss and delay
and their effects on real-time multimedia service quality'. ACM Network
and Operating Systems Support for Digital Audio and Video.
 Keogh, E., Lin, J. & Fu, A. (2005), Hot SAX: Efficiently finding the most
unusual time series subsequence, in `Fifth IEEE International
Conference on Data Mining (ICDM'05)', IEEE, Houston, pp. 1-8. Fifth
IEEE International Conference on Data Mining.
27 September 2013 24
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
References
 Lau, H. Y. K. & Woo, S. O. (2007), `An agent-based dynamic routing
strategy for automated material handling systems', International Journal
of Computer Integrated Manufacturing 21(3), 269-288.
 Miloucheva, I., Hofmann, U. & Gutirrez, P. A. A. (2003), Spatio-temporal
QoS pattern analysis in large scale internet environment, in G. Ventre &
R. Canonico, eds, `Interactive Multimedia on Next Generation Networks,
LNCS', Vol. 2899, Springer, Berlin, p. 282. Interactive Multimedia on
Next Generation Networks: First International Work shop on Multimedia
Interactive Protocols and Systems, MIPS 2003, Napoli, Italy, November
2003: Proceedings.
 Moon, S. B. (2000), Measurement and analysis of end-to-end delay and
loss in the Internet, PhD thesis, University of Massachusetts, Amherst.
 Paxson, V. (1998), `Measurements and analysis of end-to-end internet
dynamics', University of California at Berkeley, Berkeley, CA .
 Rankin, J., Christie, G. & Kondratova, I. (2005), Mobile multimodal
solutions for project closeout, in `Proceedings of the CSCE 6th
Construction Specialty Conference,', Toronto, Ontario, pp. 2-4.
 Rocha-Mier, L. E., Sheremetov, L. & Batyrshin, I. (2007), `Intelligent
agents for real time data mining in telecommunications networks',
Lecture Notes in Computer Science 4476, 138.
27 September 2013 25
The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining
(PAKDD2013)
References
 Roychoudhuri, L. & Al-Shaer, E. (2005), `Real-time packet loss
prediction based on end-to-end delay variation', IEEE Trans. Network
Service Manager 2(1).
 Tobe, Y., Tamura, Y., Molano, A., Ghosh, S. & Tokuda, H. (2000),
Achieving moderate fairness for UDP ows by path-status classification,
in `Local Computer Networks, 2000. LCN 2000. Proceedings. 25th
Annual IEEE Conference on', pp. 252-261.
 Wang, Z. & Crowcroft, J. (1991), `A new congestion control scheme:
slow start and search (Tri-S)', SIGCOMM Comput. Commun. Rev. 21(1),
32-43. 116033.
 Weiss, G., Eddy, J., Weiss, S. & Dube, R. (1998), Intelligent
telecommunication technologies, in L. C. Jain, R. Johnson, Y. Takefuji &
L. Zadeh, eds, `Knowledge-Based Intelligent Techniques in Industry',
CRC Press, Boca Raton, pp. 249-275.
 Zadeh, L. A. (2001), `From computing with numbers to computing with
words from manipulation of measurements to manipulation of
perceptions', Annals of the New York Academy of Sciences 929(1), 221-
252.
27 September 2013 26

More Related Content

Viewers also liked

Istoriy shkoli 1894 god
Istoriy shkoli  1894 godIstoriy shkoli  1894 god
Istoriy shkoli 1894 godhimbaza
 
IQS logo
IQS logoIQS logo
IQS logoSigur
 
Digital Imaging Narrative
Digital Imaging NarrativeDigital Imaging Narrative
Digital Imaging Narrativedunhamas
 
доклад
докладдоклад
докладau-elista
 
OECD perspective on social innovation and the public sector
OECD perspective on social innovation and the public sectorOECD perspective on social innovation and the public sector
OECD perspective on social innovation and the public sectorESF Vlaanderen
 
Expresiones en español con la palabra gato
Expresiones en español con la palabra gatoExpresiones en español con la palabra gato
Expresiones en español con la palabra gatoEuroace
 

Viewers also liked (16)

Istoriy shkoli 1894 god
Istoriy shkoli  1894 godIstoriy shkoli  1894 god
Istoriy shkoli 1894 god
 
Abc egypted
Abc egyptedAbc egypted
Abc egypted
 
IQS logo
IQS logoIQS logo
IQS logo
 
Digital Imaging Narrative
Digital Imaging NarrativeDigital Imaging Narrative
Digital Imaging Narrative
 
Master copy
Master copyMaster copy
Master copy
 
20 de junio
20 de junio20 de junio
20 de junio
 
Trabajo de lengua
Trabajo de lenguaTrabajo de lengua
Trabajo de lengua
 
Redes
RedesRedes
Redes
 
Invitation bw 2014
Invitation bw 2014Invitation bw 2014
Invitation bw 2014
 
доклад
докладдоклад
доклад
 
E learning
E learningE learning
E learning
 
UDLA 2014 - Valeria Carreño - Javiera Ortúzar
UDLA 2014 - Valeria Carreño - Javiera OrtúzarUDLA 2014 - Valeria Carreño - Javiera Ortúzar
UDLA 2014 - Valeria Carreño - Javiera Ortúzar
 
Manual project 2003
Manual project 2003Manual project 2003
Manual project 2003
 
OECD perspective on social innovation and the public sector
OECD perspective on social innovation and the public sectorOECD perspective on social innovation and the public sector
OECD perspective on social innovation and the public sector
 
Expresiones en español con la palabra gato
Expresiones en español con la palabra gatoExpresiones en español con la palabra gato
Expresiones en español con la palabra gato
 
Characterization Presentation
Characterization PresentationCharacterization Presentation
Characterization Presentation
 

Similar to PAKDD2013

Ppt for paper id 696 a review of hybrid data mining algorithm for big data mi...
Ppt for paper id 696 a review of hybrid data mining algorithm for big data mi...Ppt for paper id 696 a review of hybrid data mining algorithm for big data mi...
Ppt for paper id 696 a review of hybrid data mining algorithm for big data mi...Prasanta Paul
 
New Research Articles 2020 November Issue International Journal of Software E...
New Research Articles 2020 November Issue International Journal of Software E...New Research Articles 2020 November Issue International Journal of Software E...
New Research Articles 2020 November Issue International Journal of Software E...ijseajournal
 
Open Analytics Environment
Open Analytics EnvironmentOpen Analytics Environment
Open Analytics EnvironmentIan Foster
 
TOP 10 ROUTING PAPERS: RECOMMENDED READING – NETWORK RESEARCH
TOP 10 ROUTING PAPERS: RECOMMENDED READING – NETWORK RESEARCHTOP 10 ROUTING PAPERS: RECOMMENDED READING – NETWORK RESEARCH
TOP 10 ROUTING PAPERS: RECOMMENDED READING – NETWORK RESEARCHIJCNCJournal
 
Using Embeddings for Dynamic Diverse Summarisation in Heterogeneous Graph Str...
Using Embeddings for Dynamic Diverse Summarisation in Heterogeneous Graph Str...Using Embeddings for Dynamic Diverse Summarisation in Heterogeneous Graph Str...
Using Embeddings for Dynamic Diverse Summarisation in Heterogeneous Graph Str...Niki Pavlopoulou
 
The lifecycle of reproducible science data and what provenance has got to do ...
The lifecycle of reproducible science data and what provenance has got to do ...The lifecycle of reproducible science data and what provenance has got to do ...
The lifecycle of reproducible science data and what provenance has got to do ...Paolo Missier
 
Streaming Hypothesis Reasoning - William Smith, Jan 2016
Streaming Hypothesis Reasoning - William Smith, Jan 2016Streaming Hypothesis Reasoning - William Smith, Jan 2016
Streaming Hypothesis Reasoning - William Smith, Jan 2016Seattle DAML meetup
 
Federated Learning of Neural Network Models with Heterogeneous Structures.pdf
Federated Learning of Neural Network Models with Heterogeneous Structures.pdfFederated Learning of Neural Network Models with Heterogeneous Structures.pdf
Federated Learning of Neural Network Models with Heterogeneous Structures.pdfKundjanasith Thonglek
 
June 2020: Most Downloaded Article in Soft Computing
June 2020: Most Downloaded Article in Soft Computing  June 2020: Most Downloaded Article in Soft Computing
June 2020: Most Downloaded Article in Soft Computing ijsc
 
DISTRILAKE PROJECT D1 module 20_3_2013
DISTRILAKE PROJECT D1 module 20_3_2013DISTRILAKE PROJECT D1 module 20_3_2013
DISTRILAKE PROJECT D1 module 20_3_2013Carolina Arias Muñoz
 
ArrayUDF: User-Defined Scientific Data Analysis on Arrays
ArrayUDF: User-Defined Scientific Data Analysis on ArraysArrayUDF: User-Defined Scientific Data Analysis on Arrays
ArrayUDF: User-Defined Scientific Data Analysis on ArraysGoon83
 
Iciic 2010 114
Iciic 2010 114Iciic 2010 114
Iciic 2010 114hanums1
 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...ijceronline
 
Generalized Notions of Data Depth
Generalized Notions of Data DepthGeneralized Notions of Data Depth
Generalized Notions of Data DepthMukund Raj
 
ACT Talk, Giuseppe Totaro: High Performance Computing for Distributed Indexin...
ACT Talk, Giuseppe Totaro: High Performance Computing for Distributed Indexin...ACT Talk, Giuseppe Totaro: High Performance Computing for Distributed Indexin...
ACT Talk, Giuseppe Totaro: High Performance Computing for Distributed Indexin...Advanced-Concepts-Team
 

Similar to PAKDD2013 (20)

Enabling semantic integration
Enabling semantic integration Enabling semantic integration
Enabling semantic integration
 
Ppt for paper id 696 a review of hybrid data mining algorithm for big data mi...
Ppt for paper id 696 a review of hybrid data mining algorithm for big data mi...Ppt for paper id 696 a review of hybrid data mining algorithm for big data mi...
Ppt for paper id 696 a review of hybrid data mining algorithm for big data mi...
 
Seminar
SeminarSeminar
Seminar
 
New Research Articles 2020 November Issue International Journal of Software E...
New Research Articles 2020 November Issue International Journal of Software E...New Research Articles 2020 November Issue International Journal of Software E...
New Research Articles 2020 November Issue International Journal of Software E...
 
Open Analytics Environment
Open Analytics EnvironmentOpen Analytics Environment
Open Analytics Environment
 
TOP 10 ROUTING PAPERS: RECOMMENDED READING – NETWORK RESEARCH
TOP 10 ROUTING PAPERS: RECOMMENDED READING – NETWORK RESEARCHTOP 10 ROUTING PAPERS: RECOMMENDED READING – NETWORK RESEARCH
TOP 10 ROUTING PAPERS: RECOMMENDED READING – NETWORK RESEARCH
 
Using Embeddings for Dynamic Diverse Summarisation in Heterogeneous Graph Str...
Using Embeddings for Dynamic Diverse Summarisation in Heterogeneous Graph Str...Using Embeddings for Dynamic Diverse Summarisation in Heterogeneous Graph Str...
Using Embeddings for Dynamic Diverse Summarisation in Heterogeneous Graph Str...
 
The lifecycle of reproducible science data and what provenance has got to do ...
The lifecycle of reproducible science data and what provenance has got to do ...The lifecycle of reproducible science data and what provenance has got to do ...
The lifecycle of reproducible science data and what provenance has got to do ...
 
Streaming Hypothesis Reasoning - William Smith, Jan 2016
Streaming Hypothesis Reasoning - William Smith, Jan 2016Streaming Hypothesis Reasoning - William Smith, Jan 2016
Streaming Hypothesis Reasoning - William Smith, Jan 2016
 
Federated Learning of Neural Network Models with Heterogeneous Structures.pdf
Federated Learning of Neural Network Models with Heterogeneous Structures.pdfFederated Learning of Neural Network Models with Heterogeneous Structures.pdf
Federated Learning of Neural Network Models with Heterogeneous Structures.pdf
 
ME Synopsis
ME SynopsisME Synopsis
ME Synopsis
 
June 2020: Most Downloaded Article in Soft Computing
June 2020: Most Downloaded Article in Soft Computing  June 2020: Most Downloaded Article in Soft Computing
June 2020: Most Downloaded Article in Soft Computing
 
DISTRILAKE PROJECT D1 module 20_3_2013
DISTRILAKE PROJECT D1 module 20_3_2013DISTRILAKE PROJECT D1 module 20_3_2013
DISTRILAKE PROJECT D1 module 20_3_2013
 
ArrayUDF: User-Defined Scientific Data Analysis on Arrays
ArrayUDF: User-Defined Scientific Data Analysis on ArraysArrayUDF: User-Defined Scientific Data Analysis on Arrays
ArrayUDF: User-Defined Scientific Data Analysis on Arrays
 
Iciic 2010 114
Iciic 2010 114Iciic 2010 114
Iciic 2010 114
 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
 
SSG4Env EGU2010
SSG4Env EGU2010SSG4Env EGU2010
SSG4Env EGU2010
 
Generalized Notions of Data Depth
Generalized Notions of Data DepthGeneralized Notions of Data Depth
Generalized Notions of Data Depth
 
Satellite Image Data Service
Satellite Image Data ServiceSatellite Image Data Service
Satellite Image Data Service
 
ACT Talk, Giuseppe Totaro: High Performance Computing for Distributed Indexin...
ACT Talk, Giuseppe Totaro: High Performance Computing for Distributed Indexin...ACT Talk, Giuseppe Totaro: High Performance Computing for Distributed Indexin...
ACT Talk, Giuseppe Totaro: High Performance Computing for Distributed Indexin...
 

Recently uploaded

Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhisoniya singh
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitecturePixlogix Infotech
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxOnBoard
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Servicegiselly40
 
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...gurkirankumar98700
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024Results
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...HostedbyConfluent
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsMaria Levchenko
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesSinan KOZAK
 

Recently uploaded (20)

Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Service
 
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
Kalyanpur ) Call Girls in Lucknow Finest Escorts Service 🍸 8923113531 🎰 Avail...
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen Frames
 

PAKDD2013

  • 1. Hooman Homayounfard, Paul Kennedy & Robin Braun Faculty of Engineering & Information Technology University of Technology, Sydney {hooman.homayounfard, paul.kennedy, Robin.Braun}@uts.edu.au NARGES: Loss prediction model based on symbolic delay forecasting 27 September 2013
  • 2. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Agenda  Introduction  Background  NARGES Model  Model Evaluation  Results  Conclusions 227 September 2013
  • 3. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Introduction: Internet FISH Problem 327 September 2013 Can we risk by sending valuable data stream to a highly congested “Hi-Performance” link? AS 2AS 1 Internet Cloud
  • 4. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Introduction  Quality of Service (QoS) time-series associated to a link within the Internet  The “time cost” of numerical time-series analysis methods for “trivial forecasting accuracy” is prohibitive.  Researchers suggest a transition from computing with numbers to the manipulation of perceptions in the form of linguistic variables (Zadeh, 2001; Keogh et al., 2005; Batyrshin & Sheremetov, 2008). 427 September 2013
  • 5. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Introduction: Quality of Service (QoS)  Related network parameters that have impact on the QoS of the audio/video applications Delay Jitter Packet-loss  There is correlation between delay (and jitter) values and the probability of packet-loss occurrence in a link. (Roychoudhuri 2007) 527 September 2013
  • 6. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013)  “Computing with words as a transformation of an initial data set (IDS) into a terminal dataset (TDS)” (Zadeh ,1999 )  Computing with Words (CW): p and q are prepositions stated in human linguistic terms Background (I) 6 CWIDS TDS Initial data set: (p) Terminal data set: (q) 27 September 2013
  • 7. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Background (II) 7  Let X, Y be independent random variables taking values in a finite set of linguistic terms V = {v1,v2,v3,…,vn} with probabilities p1,…, pn and q1,…, qn, respectively. 27 September 2013
  • 8. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) NARGES Model Overview 827 September 2013 Forecasted Delay Forecasted Jitter Forecasted Delay Forecasted Delay & Jitters (t) Packet-Loss Time Series Data Network (used for training MLP ANN) Measured QoS Time Series (t-1 & t-2) Forecasting Module (HDAX) Pattern Classification & Approximation Pattern Database/Look-up Table of triplet i-j-k Trend Patterns and their observed frequencies Pattern Frequencies Retrieval for Approximation Storing Patterns & Frequencies PDB Predicted Packet-Loss (t+1) Delayed Incoming QoS Data (Delay, Jitter, Packet-loss) MLP ANN (Predictive Module) Predicted Packet-Loss Forecasted Jitter Measured Jitter Measured Delay  NARGES cosist of two subcomponent:  Forecasting Module (HDAX)  Predictive Module (MLP)
  • 9. Forecasting Module (HDAX) Forecasted Delay Forecasted Delay & Jitters (t) Measured QoS Time Series (t-1 & t-2) Forecasting Module (HDAX) Pattern Classification & Approximation Pattern Database/Look-up Table of triplet i-j-k Trend Patterns and their observed frequencies Pattern Frequencies Retrieval for Approximation Storing Patterns & Frequencies PDB Delayed Incoming QoS Data (Delay, Jitter, Packet-loss) Forecasted Jitter Measured Jitter Measured Delay 927 September 2013
  • 10. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) HDAX: Deterministic Mapping Function (DMF) Basic Patterns are described in terms of linguistic variables by Deterministic mapping Function (DMF) like Alphabet = {SI, I, P, D, SD, OUT}  DMF: The scale of time-series trends 27 September 2013 10
  • 11. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) HDAX: Basic & Composite Patterns  Pattern Definition a. Basic Pattern b. Composite Pattern 27 September 2013
  • 12. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013)  Transition from Number to Linguistic Values HDAX: Transition from Number to Linguistic Values 12 SI SI SD SDSI t Y Time Values f : a numerical (crisp) function f *: approximation of f If Time is t then Y is Yt Yt is a Linguistic value assigned to a Y numerical value at time t D D P P SI 27 September 2013
  • 13. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) 6000 - - 5000 - - 4000 - - 3000 - - 2000 - - 1000 - - Y-Axis Time SD(4) I (1) D (3) P (0) I (1) P (0) SI (2) I (1) SI (2) SD (4) SI (2) I (1) D (3) Time-SeriesValues End-to-endDelay/Jitter SD (4) I – D – P P – SI – I Pattern Database (Runtime: Lookup Table) . . . . . . . . . P – P – P SD – SD – SD . . . Y-Axis yt-2 yt-1 yt yt-3 HDAX: Training & Simulation 13 Previous(i) Current(j) Next(k) • Training time: The i − j − k patterns are counted to populate the corresponding i − j − k frequency entries in the pattern database. • Simulation time: •The frequencies stored in Pattern Database (Look-up Table) is used for finding the most frequent pattern having previous two time-series trends. • This estimated trend is used to approximate the yt values using Euler method. 27 September 2013
  • 14. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Predictive Module (MLP) 1427 September 2013 Forecasted Jitter Forecasted Delay Forecasted Delay & Jitters (t) Predicted Packet-Loss (t+1) MLP ANN (Predictive Module) Predicted Packet-Loss Measured QoS Time Series (t-1 & t-2) Delayed Incoming QoS Data (Delay, Jitter, Packet-loss) Measured Packet-Loss Time Series (used for training MLP ANN)
  • 15. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013)  Parameter settings is done based on the optimum MLP results of  Loss Function (NRMSE)  Performance (speed)  MLP Training The Levenberg-Marquardt optimisation process is based upon Bayesian regularization that minimises:  MSE, and  weights, MLP: Parameter Setting & Training 1527 September 2013
  • 16. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Data Sources (I)  Datasets:  Archives are provided by Napoli University “Federico II” generated by D-ITG Traffic Generator http://www.grid.unina.it/Traffic/Traces/dtraces (36 Datasets)  Network logs simulated on OPNET Modeller http://www.opnet.com/solutions/network_rd/modeler .html (10 Datasets)  Data used for:  Training (25%)  Simulation (75%). 1627 September 2013
  • 17. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Model Evaluation  Evaluation Benchmark: Autoregressive Moving Average (ARMA)  Measurements  Error Measurement (Loss Function): Normalised Root Mean Square Error (NRMSE) Speed Measurement  Cross-Correlation Function (CCF)  Non-parametric Holm’s test used to interpret the results. We Tested the similarity of algorithms for accuracy, speed and cross-correlation (CCF) over Delay, Jitter and Packet-Loss 1727 September 2013
  • 18. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Results: Experiments D-ITG Datasets  Holm Table for α = 0.05. Models printed in bold are statistically significantly better. 1827 September 2013
  • 19. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Results: Experiments OPNET Datasets  Holm Table for = 0.05. Models printed in bold are statistically significantly better. 1927 September 2013
  • 20. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Conclusions & Future Work  Conclusions  NARGES results were statistically shown to have significant correlation and precision compared to the measured values on a link.  The results demonstrated the quality and precision improvement of NARGES compared to ARMA model.  Future Work  Implementing the model within a co-simulation environment for online model evaluation.  Info. on: www-staff.it.uts.edu.au/~hoomanhm 2027 September 2013
  • 21. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) Thanks for your attention Hooman Homayounfard Hooman.Homayounfard@uts.edu.au 27 September 2013 21
  • 22. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) References  Batyrshin, I. Z. & Sheremetov, L. B. (2008), `Perception-based approach to time series data mining', Applied Soft Computing Journal 8(3),1211- 1221.  Biaz, S. & Vaidya, N. H. (1998), Distinguishing congestion losses from wireless transmission losses: A negative result, in `7th Int. Conf. Computer Communications and Networks', Lafayette, LA, pp. 722-731. Proceedings of the Seventh International Conference on Computer Communications and Networks (IC3N).  Botta, A., Dainotti, A. & Pescap, A. (2007), Multi-protocol and multi- platform traffic generation and measurement, in `IEEE INFOCOM 2007', Anchor-age, Alaska, USA.  Box, G. E. P., Jenkins, G. M. & Reinsel, G. C. (2008), Time series analysis: forecasting and control, Wiley series in probability and statistics, 4th edn, John Wiley, Hoboken, N.J.  Brakmo, L. S., O'Malley, S. W. & Peterson, L. L. (1994), `TCP Vegas: New techniques for congestion detection and avoidance', ACM SIGCOMM Computer Communication Review 24(4), 24-35.  Bui, V., Zhu, W., Pescape, A. & Botta, A. (2007), Long horizon end-to- end delay forecasts: A multi-step-ahead hybrid approach, in `12th IEEE Symposium on Computers and Communications, 2007. ISCC 2007', IEEE, pp. 825-832. 27 September 2013 22
  • 23. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) References  Carter, R. L. & Crovella, M. E. (1996), Measuring bottleneck link speed in packet-switched networks, Vol. 27, 28 of Performance evaluation, Boston University, Boston, MA, USA.  Chatfield, C. (2001), Time-series forecasting, Chap-man & Hall.  Debenham, J. & Simoff, S. (2007), Informed agents: Integrating data mining and agency, in C. Boukis, L. Pnevmatikakis & L. Polymenakos, eds, `International Federation for Information Processing-IFIP', Vol. 247, Springer-Verlag, Boston, pp. 165-173.  Dovrolis, C., Ramanathan, P. & Moore, D. (2001), What do packet dispersion techniques measure?, in `INFOCOM 2001. Twentieth Annual Joint Conference of the IEEE Computer and Communications Societies. Proceedings. IEEE', Vol. 2, pp. 905-914.  Handley, M. (1997), An examination of MBone performance, Technical report, ISI Research Report ISI/RR-97.  Hannan, E. & Rissanen, J. (1982), `Recursive estimation of mixed autoregressive-moving average order', Biometrika 69(1), 81-94. 27 September 2013 23
  • 24. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) References  Hatonen, K., Klemettinen, M., Mannila, H., Ronkainen, P. & Toivonen, H. (1996), TASA: Telecommunication alarm sequence analyzer or how to enjoy faults in your network, in `Network Operations and Management Symposium, 1996., IEEE', Vol. 2, pp. 520-529.  Hermanns, O. & Schuba, M. (1996), `Performance investigations of the IP multicast architecture', Computer Networks and ISDN Systems 28(4), 429-439.  Jain, M. & Dovrolis, C. (2002), End-to-end available bandwidth: Measurement methodology, dynamics, and relation with TCP throughput, in `Proceedings of the 2002 SIGCOMM conference 4', Vol. 32, ACM New York, NY, USA, pp. 295-308.  Jain, R. (1989), `A delay-based approach for congestion avoidance in interconnected heterogeneous computer networks', SIGCOMM Comput. Commun. Rev. 19(5), 56-71. 74686.  Jiang, W. & Schulzrinne, H. (2000), `Modeling of packet loss and delay and their effects on real-time multimedia service quality'. ACM Network and Operating Systems Support for Digital Audio and Video.  Keogh, E., Lin, J. & Fu, A. (2005), Hot SAX: Efficiently finding the most unusual time series subsequence, in `Fifth IEEE International Conference on Data Mining (ICDM'05)', IEEE, Houston, pp. 1-8. Fifth IEEE International Conference on Data Mining. 27 September 2013 24
  • 25. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) References  Lau, H. Y. K. & Woo, S. O. (2007), `An agent-based dynamic routing strategy for automated material handling systems', International Journal of Computer Integrated Manufacturing 21(3), 269-288.  Miloucheva, I., Hofmann, U. & Gutirrez, P. A. A. (2003), Spatio-temporal QoS pattern analysis in large scale internet environment, in G. Ventre & R. Canonico, eds, `Interactive Multimedia on Next Generation Networks, LNCS', Vol. 2899, Springer, Berlin, p. 282. Interactive Multimedia on Next Generation Networks: First International Work shop on Multimedia Interactive Protocols and Systems, MIPS 2003, Napoli, Italy, November 2003: Proceedings.  Moon, S. B. (2000), Measurement and analysis of end-to-end delay and loss in the Internet, PhD thesis, University of Massachusetts, Amherst.  Paxson, V. (1998), `Measurements and analysis of end-to-end internet dynamics', University of California at Berkeley, Berkeley, CA .  Rankin, J., Christie, G. & Kondratova, I. (2005), Mobile multimodal solutions for project closeout, in `Proceedings of the CSCE 6th Construction Specialty Conference,', Toronto, Ontario, pp. 2-4.  Rocha-Mier, L. E., Sheremetov, L. & Batyrshin, I. (2007), `Intelligent agents for real time data mining in telecommunications networks', Lecture Notes in Computer Science 4476, 138. 27 September 2013 25
  • 26. The 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD2013) References  Roychoudhuri, L. & Al-Shaer, E. (2005), `Real-time packet loss prediction based on end-to-end delay variation', IEEE Trans. Network Service Manager 2(1).  Tobe, Y., Tamura, Y., Molano, A., Ghosh, S. & Tokuda, H. (2000), Achieving moderate fairness for UDP ows by path-status classification, in `Local Computer Networks, 2000. LCN 2000. Proceedings. 25th Annual IEEE Conference on', pp. 252-261.  Wang, Z. & Crowcroft, J. (1991), `A new congestion control scheme: slow start and search (Tri-S)', SIGCOMM Comput. Commun. Rev. 21(1), 32-43. 116033.  Weiss, G., Eddy, J., Weiss, S. & Dube, R. (1998), Intelligent telecommunication technologies, in L. C. Jain, R. Johnson, Y. Takefuji & L. Zadeh, eds, `Knowledge-Based Intelligent Techniques in Industry', CRC Press, Boca Raton, pp. 249-275.  Zadeh, L. A. (2001), `From computing with numbers to computing with words from manipulation of measurements to manipulation of perceptions', Annals of the New York Academy of Sciences 929(1), 221- 252. 27 September 2013 26