Volume 2 - Issue 2, March - April 2026

Most online retail platforms still infer customer preference mainly from overt actions such as clicks, cart additions, and purchases. Although these signals are useful, they offer only a partial view of user intent. Repeatedly ignored products, temporary shopping goals, and expired interests are often left unmodeled, which can lead to repetitive suggestions and outdated personalization. This paper introduces IntelliShop, an adaptive recommendation framework for electronic commerce that addresses these gaps through four coordinated components: User Silence Intelligence for learning from non engagement, a Self Learning Taste Reset Mechanism for reducing obsolete preferences, One Tap Mood Selection for session specific control, and Daily Interest Reset for maintaining freshness. A prototype system was implemented and tested in a pilot retail environment. The findings show better recommendation relevance, fewer recurring ignored items, and quicker response to changes in user behavior when compared with conventional recommendation approaches centered primarily on positive interaction data.

R.Swetha, Nikitha M, Agalya S,"IntelliShop A Unified AI Recommendation Framework for Adaptive Personalization in Modern Electronic Commerce" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(2): Page(1403-1411) Mar-Apr 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.