Transcript Lecture10 - The University of Texas at Dallas

Building and Analyzing
Social Networks
Web Data and Semantics in Social
Network Applications
Dr. Bhavani Thuraisingham
February 15, 2013
4/1/2016 10:08
23-2
Outline
0 Reference: P. Mika, Semantic Web and Social Networks,
0
0
0
0
Springer, 2008: Chapter 3, 4, 5, 6
Electronic Sources for Network Analysis
Knowledge Representation on the Semantic Web
Modeling and Aggregating Social Network Data
Developing Social Semantic Applications
4/1/2016 10:08
23-3
Electronic Sources for Network Analysis
0 Electronic Discussion Networks
0 Blogs and Online Communications
0 Web-based Networks
4/1/2016 10:08
23-4
Electronic Discussion Networks
0 Communication among employees using email archive
0 Email networks
- E.g., Enron email network analysis
0 Build network from the email communications
0 Public forums and email lists
0 Group communication
4/1/2016 10:08
23-5
Blogs and Online Communications
0 Content analysis of blogs (web logs)
0 Trend analysis of blogs
0 Online social networks
- Facebook, Twitter, LinkedIn, Foursquare
0 Sentiment analysis
4/1/2016 10:08
23-6
Web-based Networks
0 Web pages from a network
0 Contents of web pages
0 Mine and analyze the web pages
0 Web Mining
- Web content mining
- Web structure mining
- Web log mining (who visited the web pages)
4/1/2016 10:08
23-7
Knowledge Representation on the Semantic Web
0 Ontologies and their role in the semantic web
0 Ontology languages for the semantic web
4/1/2016 10:08
23-8
Ontologies and Their Role in the Semantic Web
0 Ontologies are expressed in formal languages with well-
defined semantics
0 Ontologies build upon a shared understanding with a
community
0 RDF and OWL are languages for the semantic web
0 More expressive languages have less reasoning power
4/1/2016 10:08
23-9
Ontology Languages for the Semantic Web
0 RDF
0 RDF Schema
0 RDF Vocabulary
0 RDF and FOAF
0 RDF and Semantics
0 SPARQL (query language for RDF)
0 OWL – Web Ontology Language
0 Comparison to UML and the ER Model
4/1/2016 10:08
23-10
Modeling and Aggregating Social Network Data
0 Network Data Representation
0 Ontological Representation of Social Individuals
0 Ontological Relationship of Social Relationships
0 Aggregating and Reasoning with Social Network Data
4/1/2016 10:08
23-11
Network Data Representation
0 Graphs
0 Matrices
0 Number the nodes and use the numbers to represent the
edges (e.g., 12 means edge between nodes 1 and 2)
0 GraphML (XML for graphs)
0 Do not support the aggregation of network data
0 Key challenges: Identification and Disambiguation
4/1/2016 10:08
23-12
Ontological Representation of Social Individuals
0 FOAF is an example of an ontological representation of
individuals
0 Eliminates the drawbacks of early social networks like
Friendster, Orkut
0 The early social networks had centralized control and were
difficult to manage
0 FOAF is distributed and has a rich ontology to characterize
individuals
4/1/2016 10:08
23-13
Ontological Representation of Social
Relationships
0 Social networks such as FOAF need to be extended to
0
0
0
0
support relationships
Support the integration of social information
Integrates/aggregates multiple social networks
Properties of relationships
- Sign: Positive or Negative relationships
- Strength (e.g., frequency of contact)
- Provenance (different ways of viewing relationships)
- Relationship History
- Relationship roles
Conceptual models for social data – semantic net, RDF
4/1/2016 10:08
23-14
Aggregating and Reasoning with Social Network
Data
0 Representing Identity
- URI (Universal Resource Identifier)
- Disambiguation (A and B are the same; There are two
people called John Smith)
- OWL has the “sameAS” property
0 Equality
0 The property sameAs is reflexive, symmetric and transitive
0 Descriptive Logic vs. Rule based reasoners
- Rule based reasoners use forward chaining and backward
chaining
- Descriptive logic is used for classification and checking
for ontology consistency
4/1/2016 10:08
23-15
Developing Social Semantic Applications
0 Building Semantic Web Applications with Social Network
Features
0 Flink: The Social Network of the Semantic Web Community
0 Openacademia: Distributed semantic web-based publication
management
4/1/2016 10:08
23-16
Building Semantic Web Applications with Social
Network Data
0 General Architecture
- Sesame for storage and reasoning (alternative is Jena)
= Sesame manages the data sources
- Sesame Client API
- Querying through SPARQL
- Elmo and associated tools for building ontologies and
interfacing to RDF data
0 Social Network Applications (e.g., FLINK) are built on top of
the architecture as applications
4/1/2016 10:08
23-17
Flink: The Social Network of the Semantic Web
Community
0 Flink was developed by Peter Mika; it is a semantic web
representation of any online social data
0 Current instantiation uses semantic web researchers are
nodes and their collaboration as links
0 Visualization tools for visualizing the nodes and links
0 Flink social networks are decomposed and stored as RDF
triples and managed by Sesame
4/1/2016 10:08
23-18
Openacademia: Distributed Semantic Web-based
Publication Management
0 Openacademia is a social network application for maintaining
scientific publications
0 Data from multiple data stores (e.g., FOAF profiles,
publications) and access via Elmo crawler
0 Data converted into RDF and managed by Sesame
0 Openacademia servlet queries Sesame (SPARQL queries) and
aggregates the data and presents to the user