Open AccessERS: Latent Dirichlet Allocation Based E-Commerce Recommendation System Using Deep Neural Network
Author(s) -
S. Sheeja R. Preethi
Publication year - 2021
Publication title -
psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.112
H-Index - 10
ISSN - 0033-3077
DOI - 10.17762/pae.v58i2.2038
Subject(s) - recommender system , latent dirichlet allocation , computer science , ambiguity , redundancy (engineering) , product (mathematics) , purchasing , artificial intelligence , world wide web , information retrieval , topic model , marketing , business , geometry , mathematics , programming language , operating system
Online shopping is the best solution for busy life in today’s world. In the past decade, there had been a massive change in the way of customer’s shopping. In days individuals can help by keeping their time and effort by purchasing pieces online with various websites. Their particular working experience using a system is present in reviews/feedback given for these products. Recommender’s solutions offer heart-felt selections so that people can end-users just by recording their valuable pursuits together with preferences. Recommender devices provide fantastic opportunities to organizations; therefore, establishing new recommender system strategies and strategies has been increasing the focus. In this document, the ERS (Ecommerce recommendation system) using a serious neural community is recommended. We tend to produce some glossary about capabilities from each product or service kind and examine these to get rid of unimportant phrases by using Latent Dirichlet Allocation. Afterward, people being used any full nerve system network having the capacity to extract rich features inside reviews-characteristics matrix to deal with sparsity, ambiguity, apart from redundancy. Most people have applied factorization because of the collaborative, integrated choice supply regulations.