http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
XGBoost와 Word2Vec을 이용한 온라인 쇼핑 패턴 기반 하이브리드 협업 필터링
박세준(Se-Joon Park),성도현(Do-Hyun Soung),변영철(Yung-Cheol Byun) 한국정보기술학회 2020 한국정보기술학회논문지 Vol.18 No.9
From the huge amount of information accumulated with the development of Internet services and the increase of users, users want quickly the information that they need. Users want to reduce wasted time on unnecessary information and are satisfied with getting the information they need. Therefore, the Internet service provided collaborative filtering, a service that recommends desired information to users. In this paper, the user recognizes the click history before purchasing an item as a shopping pattern, applies it to Word2Vec, and then trains it in XGBoost, a machine learning model, to compare the effect of Word2Vec on the learning result. When learning without applying Word2Vec, the recommendation accuracy is 83.7%, and when learning by applying Word2Vec, the recommendation accuracy is 85.8%, indicating that Word2Vec affects the improvement of recommendation accuracy.