This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
|
||||||||
|
Paper Details
Paper Title
Comparative Study of Different Similarity Functions in Collaborative Filtering
Authors
  Bharti Sharma,  Gunjan
Abstract
Recommendation System is an information filtering technique, which provides users with information, which they may be interested in. It serves as a guide for a user to navigate through a large dataset helping them in discovering products of their interest. In the last two decades, more than 250 research articles were published about research-paper recommendation systems. On one hand, it is interesting to have many recommendation techniques to produce an output as each has its own advantages but it also makes it strenuous to choose the right fit to address each scenario. Also it is challenging to turn these recommendation techniques to be used in real world situations. So, it is important to help the designing team choose the best technique to get optimal solutions in order to improve serviceability and make it cost efficient. In this paper, we here list out all the techniques i.e. the similarity functions used in collaborative filtering for ready reference and propose genetic function for genetic recommend generating method. On a dataset from MovieLens and different collaborative filtering approaches, experiments were performed, in order to evaluate proposal. Top three recommendations are shown using different similarity functions in collaborative filtering approach.
Keywords- Recommendation system, collaborating system techniques, similarity functions, genetic algorithm
Publication Details
Unique Identification Number - IJEDR2004001Page Number(s) - 1-5Pubished in - Volume 8 | Issue 4 | October 2020DOI (Digital Object Identifier) -    Publisher - IJEDR (ISSN - 2321-9939)
Cite this Article
  Bharti Sharma,  Gunjan,   "Comparative Study of Different Similarity Functions in Collaborative Filtering", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.8, Issue 4, pp.1-5, October 2020, Available at :http://www.ijedr.org/papers/IJEDR2004001.pdf
Article Preview
|
|
||||||
|