※ 기관로그인 시 무료 이용이 가능합니다.
※ 학술발표대회집, 워크숍 자료집 중 4페이지 이내 논문은 '요약'만 제공되는 경우가 있으니, 구매 전에 간행물명, 페이지 수 확인 부탁 드립니다.
3,000원
원문정보
초록
영어
A recommendation system that predicts customer preferences and recommends products to users is used in various fields. Most recommendation systems often use an overall evaluation of a rating or review, and if only an overall evaluation of a review is used, there is a limit to reflecting the detailed attributes of the review data. Therefore, this study builds a customer-attribute matrix by emotionally analyzing reviews left by customers by attribute. We then use optimization techniques to calculate the weights of each matrix and combine them into one matrix to propose a model that utilizes the results of extracting customer-to-customer similarities for collaborative filtering.