Probabilistic matrix factorization based collaborative filtering with implicit trust derived from review ratings information

Ercan, Eda
Recommender systems aim to suggest relevant items that are likely to be of interest to the users using a variety of information resources such as user profiles, trust information and users past predictions. However, typical recommender systems suffer from poor scalability, generating incomprehensible and not useful recommendations and data sparsity problem. In this work, we have proposed a probabilistic matrix factorization based local trust boosted recommendation system which handles data sparsity, scalability and understandability problems. The method utilizes the implicit trust in the review ratings of users. The experiments conducted on dataset showed that our method compares favorably with the methods in the literature. In the scope of this work, we have analyzed the effect of latent vector initialization in matrix factorization models; different techniques are compared with the selected evaluation criteria.


A Multi-objective recommendation system
Özsoy, Makbule Gülçin; Polat, Faruk; Alhajj, Reda; Department of Computer Engineering (2016)
Recommendation systems suggest items to the user by estimating their preferences. Most of the recommendation systems are based on single criterion, such that they evaluate items based on overall rating. In order to give more accurate recommendations, a recommendation system can take advantage of considering multiple criteria. Beside combining multiple criteria from a single data source, multiple criteria from multiple data sources can be combined. Recommendation methods can also be used in various applicati...
Extending singular value decomposition based recommendation systems with tags and ontology
Turgut, Yakup; Toroslu, İsmail Hakkı; Department of Computer Engineering (2014)
Due to increase of the volume of data related to user ratings on items, in recent years, recommendation systems became very popular, which uses this data in order to rec- ommend items to users in many different domains. Singular Value Decomposition is one of the most widely studied collaborative filtering recommendation techniques. In some applications users are also allowed to enter (sometimes free) tags in addition to their ratings on items. Adding tags in addition to regular users’ ratings on items have a...
An ontology-based hybrid recommendation system using semantic similarity measure and feature weighting
Ceylan, Uğur; Birtürk, Ayşe Nur; Department of Computer Engineering (2011)
The task of the recommendation systems is to recommend items that are relevant to the preferences of users. Two main approaches in recommendation systems are collaborative filtering and content-based filtering. Collaborative filtering systems have some major problems such as sparsity, scalability, new item and new user problems. In this thesis, a hybrid recommendation system that is based on content-boosted collaborative filtering approach is proposed in order to overcome sparsity and new item problems of c...
Explicit Search Result Diversification Using Score and Rank Aggregation Methods
Ozdemiray, Ahmet Murat; Altıngövde, İsmail Sengör (2015-06-01)
Search result diversification is one of the key techniques to cope with the ambiguous and underspecified information needs of web users. In the last few years, strategies that are based on the explicit knowledge of query aspects emerged as highly effective ways of diversifying search results. Our contributions in this article are two-fold. First, we extensively evaluate the performance of a state-of-the-art explicit diversification strategy and pin-point its potential weaknesses. We propose basic yet novel ...
Assessing the influence of e-commerce item recommender systems on user continuance intention for future use of recommender system
Shahmanzari, Masoud; Özkan Yıldırım, Sevgi; Department of Information Systems (2013)
In recent years, there are several research studies on initial adaptation of information systems using recommender agents. This study, however, investigates the post-adaption behavior of users of such systems. As online e-commerce service websites are attracting users, existence of a recommender technology plays a substantial role in encouraging users to continue using system by helping them to discover and find items which they may interested and subsequently prefer to purchase. Researchers found that acqu...
Citation Formats
E. Ercan, “Probabilistic matrix factorization based collaborative filtering with implicit trust derived from review ratings information,” M.S. - Master of Science, Middle East Technical University, 2010.