During the first-generation of the Web (Web 1.0) in 1990s, the focus was primarily on building the Web, and making it accessible by connecting people to online documents. In the the second-generation (Web 2.0) in 2000s, the growth of social networking sites, wikis, communication tools and folksonomies brought a new experience to our society, by connecting people to people. In addition, the emergence of the mobile Internet and mobile devices was a significant platform driving the adoption and growth of the Web. Effectively, the Web became a universal medium for data, information, and knowledge exchange. In the third-generation Web (Web 3.0), the innovation shifted toward upgrading the back-end Web infrastructure level making the Web more connected, more exposed, and more intelligent. The Web is transformed from a network of separately siloed applications and content repositories to a more seamless and interoperable as a whole. The Web now can be used to establish a new network called “Social Life Networks (SLN)”  by connecting people to real-world (life) resources for decision making at both individual and societal levels.
Bring Predictive here.
It’s how you use and interrogate that data to derive new insight and make more informed decisions that can truly help people life.
Mashing-up is a website or application that combines content from more than one source into an integrated experience The data used in a mash-up is usually accessed via an API. The goal is to bring together in a single place data sources that tend live in their own data silos. Only visual integration, and do not provide any sophisticated analysis capabilities.
Mobile apps that allow users to contribute to the society and share information to the love one in their network during the crisis situation Life 360 let family set up a private network, then with a click of a button, they can let each other know where they are and if they're safe. You can also enable background tracking so everyone in your private network can continuously share their locations with one another. The app also has a panic alert feature you can activate to immediately contact family members via text, email and a voice call to give your location at the moment you need help. There are also options for regular feature phones. For family members without a phone, there is an additional GPS device that can be provided for a fee. SOS+
Waze: crowd-sourcing which allow user to report the current situation as you can see these type of data include both space time and theme
None of them can really connect people to the real world resources based on detected situation. It is time consuming by manually browsing on the web. The comprehensive development tools and computational frameworks for effectively combining and processing these available heterogeneous streams are lacking
The data streams over the web (e.g. tweets, weather.gov feeds) are translated into a unified format and made amenable for computing in the next step. Based on application logics, multiple spatio-temporal analysis operators are formed to generate different situation recognition models. The uniform data streams are continuously fed into the models to detect and recognize real-time situations. Then, the detected situations (e.g. ‘flu outbreak’ in New England) can be combined with user parameters (e.g. ‘high temperature’, and location). Again, based on application logic, the situation based controller is constructed to send out personalized action alerts (e.g. ‘Report to CDC center on 4th street’) to each individual. In addition, the analytic reports are also available to a central analyst who can then take large-scale (state, nation, corporate, or world-wide) decisions.
There are two main components in EventShop framework which are Data Ingestor and Stream Processing Engine. A workflow of moving from heterogeneous raw data streams to actionable situations is shown in Figure 2. In the Data Ingestor component, original raw spatio-temporal data from the Web are translated into unified STT (Space-Time-Theme) format along with their numeric values using an appropriate Data Wrapper. Based on users’ defined spatio-temporal resolutions, the system aggregates each STT stream to form an E-mage stream which we will describe in more detail later in this Section. E-mage was first introduced in . These E-mage streams are then transferred to the Stream Processing Engine component for processing. Based on situation recognition model determined by the domain expert, appropriate operators are applied on the E-mage streams to detect situation. The final step is a segmentation operation that uses domain knowledge to assign appropriate class to each pixel on the E-mage. This classification results in a segmentation of an E-mage into areas characterized by the situation there. Once we know the situation, appropriate actions can be taken. Next, we will describe the challenges in building this framework.
Emage Generator: Transform data from Data Adapter into Emage representation and is responsible for both making this data directly available to the executor, as well as writing it to the disk recent buffer, and emage resolution mapper, The queries run in the executor and can access the live data directly from emage generator and the historical data from the disk. Data access method is used to handle disk overload – data reduction/ user define function etc. and create Emage stream from disk. In addition, other information such as spatial and temporal pattern, and other properties Query rewriter -> source selection
Data is continuously being collected from temporally asynchronous data streams from heterogeneous sources. Use events as unification mechanism and combine these discrete events in one schema. Spatial Temporal Thematic data streams from wearable sensors, logical sensors, and physiological sensors to collect fitness data, personal events, eating habits, sleep patterns, and every day activities. Health Persona should be viewed as a wealth of representational and computational techniques that model the person’s health and fitness related aspects in terms of multidimensional signals and applies some specific classes of operations on them to examine the co-occurrence of temporal patterns in different event categories.
We consider event as a real world occurrence that unfold over space and time.they are observable occurrences of some phenomena that have either been explicitly recorded
Use temporal information as indexing mechanism.
Note that the term event is sometimes used to refer to a large-scale activity that is unusual relative to normal patterns of behavior, such as a large meeting in a building, a malicious attack on a Web server, or a traffic accident on a freeway. ). Sometimes people use event in a different manner, by referring to individual measurements or observation called point events (e.g., the recording of a single person walking through a door at a particular time, or the activity level measurement of a single person at a particular time). However, in our work event has a different meaning. We only consider interval events, a happening that spans over a certain time interval with start time and end time. In addition to what is being said about an event (theme), where (spatial) and when (temporal) it is being said are integral components to the analysis of such data. In our work temporal aspect of events is the most important one since time is used as the indexing mechanism. As shown in the framework, we have defined four categories of event. Life event is the independent one. By all means we have (data, operations, classification, etc) we try to segment person’s life time to different segments dominated by a “life event” or a “life activity”. As person goes through his/her life activities, we try to combine observations from different data stream and create event signals to form the other three types of event categories; namely Fitness Events, Food events, and Body-Parameter Events.
get a sense of where environmental triggers may be, as well inform ways of avoiding or managing them.
Social Life Networks (Eventshop and Personal Event Shop)
Social Life Networks:
EventShop and Personal Event Shop
Experiential Systems Laboratory
University of California, Irvine
Siripen Pongpaichet, Laleh Jalali, Mengfan Tang
(Adviser: Ramesh Jain)
• Social Life Networks
• Personal EventShop
• Predictive Analytics
Connecting People to Documents
Connecting People to People
“Social Life Network”
Connecting Needs to Resources
Effectively, Efficiently, and Promptly
In given situations.
EventShop : Global Situation Detection
Evolving Global Situation
Evolving Personal Situation
Need- Resource Matcher
Internet of Things
EventShop: Recognizing Situations
from Heterogeneous Data Streams
Big Data “NEED” Big Insight
Figure Ref: http://damfoundation.org/2012/09/big-data-big-deal/
Mash Up: Google Crisis Maps
“EventShop” towards SLN
[Pongpaichet 2013] EventShop: recognizing situations in web data streams
From Heterogeneous Data Sources
to Situation Recognition
Example Notification / Alerts:
You are currently in the area where there is a high chance of flooding,
these are available shelters within 10 miles around you.
(Space, Time, Theme)
[Pongpaichet 2013] EventShop: recognizing situations in web data streams
Data Unified Format: STT
• Each data source is transformed into unified
data format, S-T-T (Space - Time - Theme)
• Then each STT data stream is aggregated to
form E-mage (Event-Image) stream
“iphone” popularity in USA
from Tweets data source
Bounding Boxes and Resolutions
• “Pyramid of E-mage” resolution
is introduced to represent the
real world in E-mage at different
• Each Stel (a pixel in the E-mage)
represents a single fixed ground
• Precision vs Computational Cost
Rasterization and Error Propagation
• Data Error Factors:
– Uncertainty of data stream
– Data loss during data aggregation
– Uncertainty during data conversion
– Data error during data conversion
• To design the most optimum situation
recognition model, we need to find the new
cost evaluation method that will consider both
data accuracy and computational cost.
Personal Event Shop:
Recognizing Evolving Situation of Person
• Understanding a person
• Personicle: Personal Chronicle
• Detecting evolving personal situation
• Recommending action
Data Data Everywhere, nor any Insight to Use.
Health Persona Framework:
Humans as Actuators.
• Event Definition:
▫ NOT abnormal observation.
▫ NOT point event.
▫ BUT interval event
Motion event Stream
Physiological event Stream
Raw Data Streams
Event Streams** Predictive
Personal Data Warehouse
* location stream, activity stream, motion stream, calendar stream
** life-event stream, motion-event stream, physiological-event stream, food-event stream
Personal EventShop Platform
Data Streams *
Learning and Inference
• Frequency Domain
• Coefficients Sum
• DC Component
• Time Domain
• Mean , Median
• Std Deviation
• Min, Max, Range
• Average Absolute
• Average Resultant Variation
Enrich Personalized Asthma Risk
• Predict air quality at air quality measuring
• Interpolate air quality at the locations not
covered by measuring sites.
• Predict personalized asthma risk by using
EventShop and Personal EventShop.
Daily Ozone Data
Xt = ji Xt-i
Personalized Asthma Risk Prediction
• The predictive model
– Correlation with Personicle
• Temporal correlation in a location
• Geo-correlation between locations
– Two sets of features
– Machine learning methods
• Supervised learning