A study on collaborative recommender system using fuzzy-multicriteria approaches

Authors:
Kandasamy Palanivel;Ramakrishnan Sivakumar
Affiliations:
Department of Computer Science, AVVM Sri Pushpam College (Autonomous), Affiliated to Bharathidasan University, Poondi, Thanjavur, Tamil Nadu, India.;Department of Computer Science, AVVM Sri Pushpam College (Autonomous), Affiliated to Bharathidasan University, Poondi, Thanjavur, Tamil Nadu, India
Venue:
International Journal of Business Information Systems
Year:
2011

Citing 0
Cited 3

Attribute-based collaborative filtering using genetic algorithm and weighted C-means algorithm

International Journal of Business Information Systems
A hybrid recommendation approach based on attributes of products using genetic algorithm and naive Bayes classifier

International Journal of Business Information Systems
An effective recommendation based on user behaviour: a hybrid of sequential pattern of user and attributes of product

International Journal of Business Information Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In collaborative recommender systems, the overall ratings on items do not provide more detail about the reason behind the user's preferences. The multicriteria ratings give details about the user's preferences in multiple aspects and provide an opportunity to compute accurate recommendations. The user ratings collected by these systems are usually subjective, imprecise and vague, because it is based on user's perceptions and opinions. Fuzzy sets are an appropriate paradigm to handle the uncertainty and fuzziness of human behaviour. Because of these reasons, we propose a collaborative recommendation approach that uses the fuzzy linguistic approach to represent multicriteria user-item preference ratings, then finds similarities using fuzzy user-based and fuzzy item-based similarity measures and computes recommendations using fuzzy aggregation-based approach. The proposed approach's performance is evaluated empirically against traditional user-based and item-based recommendation algorithms using a music recommender system developed for this research. From the evaluation results, it is observed that the proposed approach shows improvement in recommendations than the traditional algorithms.