Towards a robust solution to mitigate all content-based filtering drawbacks within a recommendation system

Authors

  • Oumaima Stitini
  • Soulaimane Kaloun
  • Omar Bencharef

DOI:

https://doi.org/10.6977/IJoSI.202309_7(7).0006

Keywords:

Recommender system, Content-based filtering, Over-specialization, Limited content, Scalability, sparsity, synonym

Abstract

Recommendation systems deliver a method to simplify the user’s desire. Recommendation systems are now commonly used on the Internet. It is helpful for suggesting items in various categories, including e-commerce, medical, education, tourism, and industrial. Electronic commerce sec- tor has taken a big place in our daily lives as an active research, which helps people find what they are looking for. This paper presents a new contribution based on the combination of different algorithms to find a suitable solution to all the drawbacks of content-based recommender systems. The main contribution of this research lies in how to solve each problem and move on to the next. This paper describes an Ideal Solution Mitigating Content Disadvantages based on Three Phases called ISMCD3P . Experiments show that the algorithm can propose an appropriate solution to solve all the problems of content-based filter- ing. Experimentations operating on real datasets are used to estimate the efficacy of our strategy.

Downloads

Published

2023-09-25