Text summarization solves the problem of presenting the information needed by a user in a compact form. There are different approaches to create well formed summaries in literature. One of the newest methods in text summarization is the Latent Semantic Analysis (LSA) method. In this thesis, different LSA based summarization algorithms are explained and two new LSA based summarization algorithms are proposed. The algorithms are evaluated on Turkish and English documents, and their performances are compared using their ROUGE scores.

No abstract

Social networks, personal blog pages, on-line transaction web-sites, expertise web pages and location based social networks provide an attractive platform for millions of users to share opinions, comments, ratings, etc. Having this kind of diverse and comprehensive information leads to difficulties for users to reach the most appropriate and reliable conclusions. Recommendation systems form one of the solutions to deal with the information overload problem by providing personalized services. Using spatial, temporal and social information on recommender systems is a recent trend that increases the performance. Also, taking into account more than one criterion can improve the performance of the recommender systems. In this paper, a location and social network aware recommender system enhanced with multi objective filtering is proposed and described. The results show that the proposed method reaches high coverage while preserving precision. Besides, the proposed method is not affected by the range of ratings and provides persistent results in different settings.

Social network platforms can use the data produced by their users to serve them better. One of the services these platforms provide is recommendation service. Recommendation systems can predict the future preferences of users using their past preferences. In the recommendation systems literature there are various techniques, such as neighborhood based methods, machine-learning based methods and matrix-factorization based methods. In this work, a set of well known methods from natural language processing domain, namely Word2Vec, is applied to recommendation systems domain. Unlike previous works that use Word2Vec for recommendation, this work uses non-textual features, the check-ins, and it recommends venues to visit/checkin to the target users. For the experiments, a Foursquare checkin dataset is used. The results show that use of continuous vector space representations of items modeled by techniques of Word2Vec is promising for making recommendations.

No abstract