•  
  •  
 

Future Computing and Informatics Journal

Future Computing and Informatics Journal

DOI

http://doi.org/10.54623/fue.fcij.5.2.3

Abstract

In this modern world of ever-increasing one-click purchases, movie bookings, music, health- care, fashion, the need for recommendations have increased the more. Google, Netflix, Spotify, Amazon and other tech giants use recommendations to customize and tailor their search engines to suit the user’s interests. Many of the existing systems are based on older algorithms which although have decent accuracies, require large training and testing datasets and with the emergence of deep learning, the accuracy of algorithms has further improved, and error rates have reduced due to the use of multiple layers. The need for large datasets has declined as well. Thus, through this research proposal propose a recommendation system based on deep learning models such as multilayer perceptron that would provide a slightly more efficient and accurate recommendations.

Share

COinS