Factors influencing recommendations for women's clothing satisfaction: A latent dirichlet allocation approach using online reviews

The rapid growth of e-commerce has transformed the way female customers shop for clothing, with an endless number of options available at their fingertips. Online reviews and product suggestions are quite important in this situation for influencing the buying decisions of women. To improve their satisfaction and optimize product offerings, e-commerce businesses need to understand the factors that influence product suggestions for business benefits. Therefore, this study investigates the factors that influence product recommendations for women's ecommerce clothing satisfaction using online reviews. The dataset consists of a varied selection of women's reviews that cover a range of clothing categories and the associated sentiments. To extract and analyze the reviews, this study used Latent Dirichlet Allocation (LDA) and natural language processing (NLP) techniques, including stemming, lemmatization, tokenization, and topic modeling. The results indicate remarkable trends. Product qualities, consumers' pleasure, and the overall purchasing experience are identified as critical factors that greatly affect product recommendations. Furthermore, the effect of various other factors was investigated on the chance of receiving positive recommendations, such as clothing categories and review lengths.