CONCLUSION & FUTURE WORK Normally, users perform search tasks using multiple applications in concert: a search engine interface presents lists of potentially relevant documents; a document reader displays their contents; and a third tool—a text editor or personal information management application—is used to record notes and assessments (MS Word and MS PowerPoint). An Interest Profile Manager infers users' interests from their interactions with the multi-applications, coupled with the characteristics of the multi-source interest modeling techniques. The resulting interest profile is used to generate visualizations that direct users' attention to documents or parts of documents that match their inferred interests. Statistical methods used in the work (tf-idf, LDA an LSA) infer the interest based on the content similarity and the Ontology-based user model infers the long-term user interest dynamically using spreading activation module. Acknowledgements : This research is supported by NSF grant Hybrid (Statistical and Knowledge-based) Multi-Application User Interest Modeling Sampath Jayarathna and Frank Shipman Computer Science & Engineering, Texas A&M University – College Station Hybrid (Statistical and Knowledge-based) Multi-Application User Interest Modeling Sampath Jayarathna and Frank Shipman Computer Science & Engineering, Texas A&M University – College Station ABSTRACT We are interested about open-ended information gathering tasks— search tasks in particular—in which people collect Web documents for interpretation and synthesis. User interests are usually distributed in different systems during search tasks. Traditional user interest modeling methods are not designed for integrating and analyzing interests from multiple sources, hence, they are not very effective for obtaining comparatively complete description of user interests in a multi- application environment. We propose an approach of user interest modeling based on multi- source interest fusion using statistical/algebraic models (tf-idf, LSA, and LDA) and knowledge-based models (Ontology). Figure 1. Multi-Application Interest Modeling and Fusion Even with the best search engine and the most effective query formulation, “search tasks” require people to work through long lists of documents to synthesize the information they need; there is usually no single document containing one right answer. In fact, as people skim early documents, they may determine additional information needs that suggest further queries and results in even more documents to process. A system can support document search tasks by recommending the documents that best match a user’s interests, thereby ensuring that the user’s time is spent efficiently on the most relevant documents. In the work we present, recommendations based on demonstrated user interest; in other words, the user’s previous interactions with the document collection, along with the characteristics of the documents, are used to infer the user’s interests. USER INTEREST MODELING WebAnnotate - During information task, useful documents may be long, and cover multiple subtopics; users may read some segments and ignore others. In order to record which portion(s) of the document pique the user’s interests, an explicit interest expressions (e.g. annotations using WebAnnotate) capturing tool is used. Latent Semantic Analysis (LSA) - The SVD based LSA can take a large matrix of term document association data and construct a semantic space where terms and documents that are closely associated can be detected with Cosine Similarity. Latent Dirichlet Allocation (LDA) - Our strategy in using LDA is to describe users as a mixture of topics and to assume that each of their actions is motivated by choosing a topic of interest and subsequently a word to describe that action from the catalog of words consistent with that particular interest. We represent each user as a bag of words extracted from those actions and we use the search task to denote generating a word from the bag. Figure 2. Personalized Search & Recommendations Ontology-based User Model - An ontological approach to user profiling has proven to be successful in addressing the cold-start problem in recommender systems where no initial information is available early on upon which to base recommendations. We model the user interests using ontological profiles by assigning implicitly derived interest scores to existing concepts in domain ontology. A spreading activation algorithm is used to maintain the interest scores based on the user’s ongoing behavior. Ontology = Long-term user interest modeling Statistical Methods = Short-term user interest modeling REFERENCES SEARCH AND RECOMMENDATIONS Our major contributions in this work can be summarized as: (1)A novel personalized search & Recommendations based on evidence coming from multiple applications and multi-source interest modeling using a hybrid of statistical and knowledge- based methods. (2)In the future we plan on creating a weighting schema to identify the importance of evidence coming from multiple applications like VKB, Web Browser, Word and PowerPoint. 1.Tolomei, G., Orlando, S. and Silvestri, F., Towards a task-based search and recommender systems. in In Proceedings of ICDE Workshops, (2010), Bae, S., Hsieh, H., Kim, D., Marshall, C.C., Meintanis, K., Moore, J.M., Zacchi, A. and Shipman, F.M. Supporting document triage via annotation-based visualizations. American Society for Information Science and Technology, 45 (1) Landauer, T., Foltz, P.W. and Laham, D. An Introduction to Latent Semantic Analysis. Discourse Processes Sieg A, Mobasher B, and Burke R., “Web search personalization with ontological user profiles,” in ACM Sixteenth Conference on Information and Knowledge Management, CIKM 2007, Lisbon, Portugal, November 2007.