点击次数：
论文类型：期刊论文
发表时间：2015-07-15
发表刊物：Journal of Computational Information Systems
收录刊物：EI、Scopus
卷号：11
期号：14
页面范围：5171-5180
ISSN号：15539105
摘要：The Web search data, not only reects the focus attention and personalized demand of users, but also contains a group of social or economic behaviors. This paper conducts a prediction model on historical sales data and Web search data. To deal with the issues that the fast fashion product's demand is highly volatile with ever-changing taste of consumers and the fast fashion product's life cycle is very short, the extreme learning machine (ELM) which tends to provide a better generalization performance and much faster learning speed has been adopted to forecast for popular trend of fast fashion products. By comparing with the traditional statistical prediction model, the prediction model proposed in this paper is more accurate after introducing the search data, and this prediction model has stronger ability to predict the inection point of the popular trend. ?, 2015, Journal of Computational Information Systems. All right reserved.