An improved content based collaborative filtering algorithm for movie recommendations
- 1 August 2017
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2017 Tenth International Conference on Contemporary Computing (IC3)
Abstract
Recommender system comprises of two prime methods which help in providing meaningful recommendations namely, Collaborative Filtering algorithm and Content-Based Filtering. In this paper, we have used a hybrid methodology which takes advantage of both Content and Collaborative filtering algorithm into account. The algorithm discussed in this article is different from the previous work in this field as it includes a novel method to find the similar content between two items. The paper incorporates an analysis that justifies this new methodology and how it can provide practical recommendations. The above approach is tested on existing user and objects data and produced improved results when compared with other two favourite methods, Pure Collaborative Filtering, and Singular Value Decomposition.Keywords
This publication has 8 references indexed in Scilit:
- Recommender systems surveyKnowledge-Based Systems, 2013
- Recommender Systems HandbookPublished by Springer Science and Business Media LLC ,2011
- A Comprehensive Survey of Neighborhood-based Recommendation MethodsPublished by Springer Science and Business Media LLC ,2010
- Solving the cold-start problem in recommender systems with social tagsEurophysics Letters, 2010
- Matrix Factorization Techniques for Recommender SystemsComputer, 2009
- Hybrid Web Recommender SystemsPublished by Springer Science and Business Media LLC ,2007
- Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performanceClimate Research, 2005
- Item-based collaborative filtering recommendation algorithmsPublished by Association for Computing Machinery (ACM) ,2001