Understanding E-consumer Online Behavior: Establishing E-commerce Performance Metrics through BERT-Light BGM Approach

This proposal draws on uses and gratifications theory to investigate how informativeness, amusement, and aggravation affect different online consumer behaviors like attitude towards the Web, Web usage, and satisfaction. The article delves into the relationship between one's attitude towards the online and its outcomes, such as web usage and contentment, as well as its causes, which include enjoyment, informativeness, and aggravation. The proposed method consists of three stages: preprocessing, feature extraction, and model training. Data cleansing and session identification are part of it. Information arrangement is another name for data pre-processing. Earth Worm Algorithm (EWA) takes advantage of a diverse population of earthworms to optimise a solution. With the help of the BERT-LightGBM, the model training procedure became more precise. The suggested method outperforms its competitors, LightGBM and BERT, with an accuracy percentage of 95.72 %.