Movie Recommendation System Based on Fuzzy Inference System and Adaptive Neuro Fuzzy Inference System

Movie Recommendation System Based on Fuzzy Inference System and Adaptive Neuro Fuzzy Inference System

Mahfuzur Rahman Siddiquee (Electrical and Computer Engineering Department, North South University, Dhaka, Bangladesh), Naimul Haider (Electrical and Computer Engineering Department, North South University, Dhaka, Bangladesh) and Rashedur M. Rahman (Electrical and Computer Engineering Department, North South University, Dhaka, Bangladesh)
Copyright: © 2015 |Pages: 39
DOI: 10.4018/IJFSA.2015100103
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

One of most prominent features that social networks or e-commerce sites now provide is recommendation of items. However, the recommendation task is challenging as high degree of accuracy is required. This paper analyzes the improvement in recommendation of movies using Fuzzy Inference System (FIS) and Adaptive Neuro Fuzzy Inference System (ANFIS). Two similarity measures have been used: one by taking account similar users' choice and the other by matching genres of similar movies rated by the user. For similarity calculation, four different techniques, namely Euclidean Distance, Manhattan Distance, Pearson Coefficient and Cosine Similarity are used. FIS and ANFIS system are used in decision making. The experiments have been carried out on Movie Lens dataset and a comparative performance analysis has been reported. Experimental results demonstrate that ANFIS outperforms FIS in most of the cases when Pearson Correlation metric is used for similarity calculation.
Article Preview

Fuzzy logic has been used in many areas that include recommendation systems, fuzzy controller, robotics etc. Recently, Semwal et al. (2015) designed fuzzy logic controller that can predict push recovery strategy for a robot. Fuzzy rules were defined in terms of roll and pitch to avoid high variability. One of the earliest implementations of recommender systems is Tapestry (Goldberg, Nichols, Oki & Terry, 1992). It depends on the opinions of people on small connected communities like office workgroups, student networks, etc. Recommender systems for large communities cannot be dependent on one another. Many other recommender systems were developed, such as Ringo (Shardanand & Maes, 1995) and Video Recommender (Hill, Stead, Rosenstein & Furnas, 1995). An issue of ACM (Resnick & Varian, 1997) discusses a number of different recommender systems.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 7: 4 Issues (2018): 1 Released, 3 Forthcoming
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing