Abstract
Users’ emotional needs can be well expressed by emotional attitude and preference in online reviews, so in this paper, we extracted emotional words from users’ reviews through establishing emotional tendency corpus for online reviews, and computed emotional scores in fuzzy algorithm. The vector for emotional tendency in online reviews was formed by Cloud Model, and recommendation was run depended on nearest neighbor by similarity computing between cloud model vectors. The experiment proves that this algorithm can provide recommendation by comparing users’ emotional tendency on items and count users’ emotional tendency in data, when scores in traditional users’ scoring matrix are few and sparse.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have