Aspect Based Text Summarization Model for the E-Commerce Recommendation System

E-commerce Recommendation System is a specialized software application designed to enhance the online shopping experience for consumers by suggesting products or services that are likely to be of interest to them. This technology leverages various algorithms and data analysis techniques to analyse user behaviour, preferences, and historical interactions with the platform. This paper aimed to construct a trustable user recommendation system based on semantic complexion through network embedding model to provide potential friend suggestions and friend request. Provenance data of users for trust computation are extracted from the timeline and chat sessions of Facebook data. The interest-based similarity between the users is computed using FCA model and semantic complexion embedding approaches. Various statistical methods are applied to evaluate the effectiveness of the similarity approaches. Based on evaluation the topic model using Latent Dirichlet Allocation (LDA) is considered for similarity computation as a part of this work. Aspect Based Text Summarization Hybrid Recommendation (ABTSHR) technique is used for recommending trusted similar users. The recommendation system groups the users into three categories with interest in Entertainment, Travel and Sports. A potential friend recommendation is made from within the trust groups evolved. The proposed method compare with parameters of precision, recall, f1-score, average, recommendation, score and click-through rate (CTR) with different groups