ISSN (online) : 2395 - 7549

 
Aim and Scope
Call for Paper
Join As Reviewer
Check Status
Submit Payment
Archives
Author Guideline
Paper Format
Publication Fee
Copyright


7



7

Similarity Measures for Recommender Systems: A Comparative Study


Author(s):

Sridhar Dilip Sondur , R. V. College of Engineering; Shantharam Nayak, R. V. College of Engineering; Amit P Chigadani, R. V. College of Engineering

Keywords:

Recommender systems, Collaborative filtering, Similarity measures

Abstract:

Recommender Systems have the ability to guide the users in a personalized way to interesting items in a large space of possible options. They have fundamental applications in e-commerce and information retrieval, providing suggestion that prune large information spaces so that users are directed towards those items that best meets the needs and preferences. A variety of approaches have been proposed but collaborative filtering has been the most popular and widely used which makes use of various similarity measures to calculate the similarity. Collaborative Filtering takes the user feedback in the form of ratings in an application area and uses it to find similarities and differences between user profiles to generate recommendations. Collaborative Filtering makes use of various similarity measures to calculate the similarity or difference between the users. This paper provides an overview on few important similarity measures that are currently being used. Different similarity measures provide different results against same input parameters. So, to understand how various similarity measures behave when they are put in different contexts but with same input, few observations are made. This paper also provides a comparison graph to help understand the results of different similarity measures.

Other Details:

Manuscript Id :J4RV2I3036
Published in :Volume : 2, Issue : 3
Publication Date: 01/06/2016
Page(s): 76-80
Download Article

News

Paper Submission

Dear Authors,
The Paper Submission for the Current Issue has started..

Submit Manuscript Online

ISSN Number:

The ISSN Number of Journal Is

2395 - 7549

Read More

Impact Factor

The Impact Factor of Journal Is

5.04

Click here

ISSN :

2395 - 7549

Frequency:

Monthly

Creative Commons License