Future Computing and Informatics Journal
Abstract
The rapid growth of e-commerce has transformed how consumers shop for fashion, including women's outfits. This study presents a comprehensive evaluation of women's outfit e-commerce models through the lens of feature selection technique. understanding customer sentiments is critical in today's marketing landscape, it provides firms with insights into how customers perceive their products and/or services and offers suggestions on how to enhance their offerings. This study aims to decipher the relationship between various characteristics in customer reviews on a women's clothing e-commerce site. Additionally, it classifies each review to determine whether it recommends the reviewed product or not, and whether the review exhibits a positive, negative, or neutral sentiment. To achieve these goals, we employed univariate and multivariate analyses on dataset attributes that includes 23,486 rows and 10 features, excluding review titles and review contents. Furthermore, we implemented the recommendation and sentiment classification model. And the experimental results show that the best performance was for the Naive Bayes classifier algorithm on the Sentiment category.
Recommended Citation
Bekhet, Hadeer Mohamed; Nasr, Mona; and ElNaggar, Rasha
(2023)
"A Reviews of Women's Outfit E-Commerce Models Using Feature Selection Techniques,"
Future Computing and Informatics Journal: Vol. 8:
Iss.
2, Article 1.
Available at:
https://digitalcommons.aaru.edu.jo/fcij/vol8/iss2/1
Included in
Business Administration, Management, and Operations Commons, E-Commerce Commons, Management Information Systems Commons