Electronic commerce refers to the buying and selling that takes place when money is sent via a computer network. The internet is utilized for purposes beyond ecommerce. For instance, people utilize it for research purposes, such as checking prices before making a purchase or learning about market updates. Online marketplaces are often regarded by reviewers as practical resources that evoke emotions, ideas, and a desire to purchase from consumers. Customer sentiment, interest, and attitude data offers feedback on a variety of topics. People are more likely to trust each other when they have similar views on
related issues, according to a number of research ideas. We postulate that when people in ecommerce networks ask for and accept feedback and ideas, a sense of reciprocal trust develops. As they made their purchases, a scientific observer noted every detail. For the purpose of reviewing e-commerce systems, we offer a sentiment similarity analysis tool that can look at things like user trust and similarity. Direct trust and trust propagation, in which one person trusts another, are the two main categories of trust. We introduce a mining method for extracting sentiment-entity word pairings’ similarity attributes. Direct trust degree is determined by sentiment similarity. To determine the trust spread, one uses the transitivity property. The shortest route is determined using the proposed trust model, and the propagation trust relationship between users is determined using an enhanced shortest path technique. To evaluate the algorithms’ performance and the models’ feasibility, a big set of ecommerce reviews is utilized. Evident from the trial’s findings, sentiment similarity analysis shows promise as a tool to enhance consumers’ faith in online retailers.