Embed Size (px)
DESCRIPTION
Citation preview
Rio de Janeiro, Brazil, 5-8 November 2011
Jair Abú Bechir Láscar Alarcó[email protected]
Otávio Calaça [email protected]
Cedric Luiz de [email protected]
IMPLEMENTING SEMANTIC QUERIES IN ONLINE SOCIAL NETWORKS
Summary
Introduction
Foundations
Airetama
Implementing Semantic Queries
Conclusions
Introduction
The Web is a great technological achievement
Access, extraction, interpretation and maintenanceinformation is done manually
There is a need for software agents that help users in finding, processing and organizing this information
Web Systems are often isolated applications. They are designed for human beings and are not suitable to be automatically accessed by software tools
1/10
Introduction
This article presents the implementation of semantic queries in online social networks
For this implementation we created a framework that uses Semantic Web to make relationships in such networks
This work was done because of the clear need for more sophisticated queries, which would enable the computer to interpret these relationships semantically
The experiment used in this article was to create a ranking of affinity between users based on common interests
2/10
Web 1.0: Internet + Hiperlinks
Web 2.0: Collaborative Web
Web 3.0: Semantic Web
Foundations
3/10
Web Services
Online Social Networks
Multi-Agent Systems
Foundations
4/10
Airetama
5/10
Airetama
SIOC provides the main concepts and properties required
to describe information from online communities
FOAF is an ontology that is used to describe people, the things they create and do,
and how they are connected
6/10
Implementing Semantic Queries
SemanticRepository
Tool Agents
ControllerAgents
MemberAgents
FacebookIntegrator
Accessing Facebook Data:
7/10
Implementing Semantic Queries
Facebook Airetama
Table/Field Class/Property Domain Range
User foaf:Person - -
User.name foaf:name owl:Thing rdfs:Literal
User.first_name foaf:firstName foaf:Person rdfs:Literal
User.last_name foaf:lastName foaf:Person rdfs:Literal
User.email foaf:mbox foaf:Agent owl:Thing
User.birthday foaf:birthday foaf:Agent rdfs:Literal
User.sex foaf:gender foaf:Agent rdfs:Literal
Page owl:Thing - -
Page_fan(page_id, uid) foaf:topic_interest foaf:Agent owl:Thing
Page.page_url foaf:homepage owl:Thing foaf:Document
Page.name foaf:name owl:Thing rdfs:Literal
Friend (uid1, uid2) foaf:knows foaf:Person foaf:Person
Converting Data from Facebook to Airetama:
8/10
Implementing Semantic Queries
Executing SPARQL from RDF Data:
9/10
Conclusions
The biggest advantage found in this work is the ability to perform semantic queries on the pre-existing data from social network users
Another contribution was the conversion of data from Facebook users to triples in RDF
It was also relevant to show that it is possible to create semantic-based social applications
Can be done an approach of potential relationships among strangers and not just a search for people you know in real life, as is done in most social networks online today
The inclusion of semantic in the current Web aims at allowing agents software to use their pages more intelligently, thus offering better services
10/10
Questions?
THANKS!
