Abstract
This paper proposes the implicative rating measure developed on the typicality measure. The paper also proposes a new recommendation model presenting the top N items to the active users. The proposed model is based on the user-based collaborative filtering approach using the implicative intensity measure to find the nearest neighbors of the active users, and the proposed measure to predict users’ ratings for items. The model is evaluated on two datasets MovieLens and CourseRegistration, and compared to some existing models such as: the item based collaborative filtering model using the Jaccard measure, the user based collaborative filtering model using Jaccard measure, the popular items based model, the latent factor based model, and the association rule based model using the confidence measure. The experimental results show that the performance of the proposed model is better when compared to other five models.
Highlights
Recommender systems/recommendation systems (RSs) [1] are techniques or software tools embedded in an application or website to predict the preferences of an individual or a group of users for a specific product or service; and/or to recommend the appropriate products or services to an individual or a group of users, thereby reducing the information overload
This paper proposes a new recommendation model based on user-based collaborative filtering and the implicative rating measure to present to the active users the top N items
We propose the implicative rating measure KnnIR (K nearest neighbors based implicative rating) to predict the ratings that can be given by the active user ua for each item i I
Summary
Recommender systems/recommendation systems (RSs) [1] are techniques or software tools embedded in an application or website to predict the preferences of an individual or a group of users for a specific product or service; and/or to recommend the appropriate products or services to an individual or a group of users, thereby reducing the information overload. The techniques (methods) of recommendation are based on the ones used in data mining and machine learning [3], [4] such as classification, clustering, association rule mining, regression models, or some of the supervised or unsupervised learning methods. Recommendation techniques are divided into two main classes: the class of basic techniques such as collaborative filtering, content filtering or hybrid; and the class of techniques developed on the basic techniques and the additional data such as the contextual information or the social information. Recommendation systems can be classified into different groups [2], [3], [5]: content based, collaborative, demographic based, knowledge based, hybrid, context based, social based, and group based. In the fields of research on recommendation systems, proposing the new recommendation models or improving the existing recommendation methods has still been the mainstream of research and received the most attention
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Advanced Computer Science and Applications
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
