A Workshop of the 22nd International Joint Conference on Artificial Intelligence (IJCAI-11)
Recent work on link mining and social networks has led to a convergence of methodologies for network modeling, incorporating classification, learning and reasoning with graphical models, frequent subgraph mining, relational representation, and link prediction, among other techniques. Many intelligent systems applications to information extraction, web search, and recommendation call for inferences to be made regarding the existence, type, or attributes of links. Some tasks, such as question answering using information networks, may require that inferences be based upon partial link information and made under uncertainty about participating entities and relationships.
The emphasis of this workshop shall be approaches based on relationship extraction from heterogeneous sources such as technical literature, news articles, social network profile data, and social media. Relevant media include, but are not limited to, forums, blogs, microblogging mechanisms such as Twitter and status updates, podcasts, video, and systems for rating and commenting. However, the scope is not limited to any particular approach to link analysis or any source of network information such as text corpora. Application areas that often exhibit a need for link analysis include:
- Bioinformatics and biomedicine: computational genomics, drug interaction, protein-protein interaction
- Content-management systems: version control, wikification
- Information security:attack graphs, information flow, mechanism design, trust networks
- Knowledge and information management: data fusion, information integration, source attribution
- Recommender systems for social networks:communities, experts, friends, products, reviewers, service providers
- Spatiotemporal reasoning:epidemiology, link formation, meme propagation
RSS 2.0