Year: 2026
Type: preprint
Abstract: E-commerce has changed from a simple online buying platform into a highly personalized digital marketplace. Every day, customers see product suggestions based on searches, clicks, purchases, ratings, browsing time and other signals. AI-based recommendation systems use these signals to predict what a customer may want next. This assignment explains the meaning, working, major types, benefits and limitations of recommendation systems in e-commerce. It also examines an important question: do humans make more mistakes than AI when making product recommendations? The answer is not simply yes or no. AI can process huge amounts of data consistently and quickly, but it can repeat biased data, misunderstand unusual situations and recommend products that are commercially useful rather than genuinely useful. Humans can be slower and less consistent, but they can understand context, emotion, common sense and ethical concerns. Therefore, the strongest e-commerce systems combine machine intelligence with human judgment.
Author: Sarvesh Pal
Institution: University of Lucknow
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7225615730