Nishtha Kukreja
Computer science
Hometown: Raipur, Chhattisgarh, India
Graduation date: Fall 2026
Additional details: Honors student
FURI | Fall 2024, Summer 2024
ESHA: Enhanced SHopping Assistance
This study aims to improve e-commerce search algorithms to enhance user experience and business performance. Traditional methods rely on standard image similarity metrics, which may not effectively capture users’ natural language expressions. By employing diffusion models and large language models (LLMs) to create vector space representations of products and user queries, we seek to significantly boost user satisfaction. Performance is measured using ranking quality metrics like hit rate. Moving beyond conventional text-based approaches, we aim to achieve a nuanced understanding of semantic content and user intent. Future research will explore integrating real-time analytics and adaptive learning to refine search capabilities.
Mentor: Sandeep Gupta