Improved probabilistic matrix factorization model for sparse datasets /

Download
index.pdf
2014
Ar, Yılmaz
Show full item record
72
views
20
downloads
The amount of information on the World Wide Web has increased significantly owing to advancing web and information technologies. This has made it difficult for users to obtain relevant and useful information thus there is a need for information filtering. Recommender Systems (RS) have emerged as a technique to overcome the problem. Collaborative Filtering (CF) that is one of the widely used RS approaches aims to predict users’ preference concerning an item. The main idea behind CF is the users who agreed in the past will agree in the future. The Probabilistic Matrix Factorization (PMF) is the preferred CF technique in the literature due to its high accuracy and scalability. This thesis demonstrates the importance of the initialization techniques for the user and the item latent vectors in the PMF algorithm with real and synthetic datasets and proposes five different initialization techniques. The suggested approaches produce better results in comparison with the state-of-the-art techniques in particularly very sparse datasets.
Databases., Database management., Sparse matrices., Information filtering systems., Recommender systems (Information filtering).
http://etd.lib.metu.edu.tr/upload/12618157/index.pdf
https://hdl.handle.net/11511/24189
Graduate School of Informatics, Thesis

Suggestions

A Comparison of different recommendation techniques for a hybrid mobile game recommender system
Cabir, Hassane Natu Hassane; Alpaslan, Ferda Nur; Çakıcı, Ruket; Department of Computer Engineering (2012)
As information continues to grow at a very fast pace, our ability to access this information effectively does not, and we are often realize how harder is getting to locate an object quickly and easily. The so-called personalization technology is one of the best solutions to this information overload problem: by automatically learning the user profile, personalized information services have the potential to offer users a more proactive and intelligent form of information access that is designed to assist us ...
A singular value decomposition approach for recommendation systems
Osmanlı, Osman Nuri; Toroslu, İsmail Hakkı; Department of Computer Engineering (2010)
Data analysis has become a very important area for both companies and researchers as a consequence of the technological developments in recent years. Companies are trying to increase their profit by analyzing the existing data about their customers and making decisions for the future according to the results of these analyses. Parallel to the need of companies, researchers are investigating different methodologies to analyze data more accurately with high performance. Recommender systems are one of the most...
Optimization of an online course with web usage mining
Akman, LE; Akkan, B; Baykal, Nazife (2004-02-18)
The huge amount of information existing in the World Wide Web constitutes an ideal environment to implement data mining techniques. Web mining is the mining of web data. There are different applications of web mining: web content mining, web structure mining and web usage mining. In our study we analyzed an online course by web usage mining techniques in order to optimize the navigation paths, the duration of the time spend on each page and the number of visits throughout the semester of the course. Moreove...
Efficient rating estimation by using similarity in multi-dimensional check-in data
Uçar, Behlül; Karagöz, Pınar; Toroslu, İ. Hakkı; Department of Computer Engineering (2014)
The usage coverage of location based social networks have boomed in the last years as well as the amount of data produced in them. This data is suitable for processing in order to make prediction. One of the requirements of this process is that the method used should be suitable for very big data sets. We propose a graph-based similarity calculation method in location-based social networks which improves the rating prediction performance of Singular Value Decomposition based collaborative filtering systems....
A content boosted collaborative filtering approach for movie recommendation based on local & global similarity and missing data prediction
Özbal, Gözde; Alpaslan, Ferda Nur; Department of Computer Engineering (2009)
Recently, it has become more and more difficult for the existing web based systems to locate or retrieve any kind of relevant information, due to the rapid growth of the World Wide Web (WWW) in terms of the information space and the amount of the users in that space. However, in today's world, many systems and approaches make it possible for the users to be guided by the recommendations that they provide about new items such as articles, news, books, music, and movies. However, a lot of traditional recommen...
Citation Formats
Y. Ar, “Improved probabilistic matrix factorization model for sparse datasets /,” Ph.D. - Doctoral Program, Middle East Technical University, 2014.
METU IIS - Integrated Information Service open@metu.edu.tr