CAJ | 학술논문

个性化推荐为解决互联网信息过载问题提供了新的思路。为有效地构建用户模型和改进个性化推荐的效果，提出了一种挖掘非结构化文本中上下文信息的新模型，将得到的上下文信息嵌入用户模型信息中，丰富了用户模型。实验结果表明，该模型应用于客户对旅馆评论的上下文数据中，能够大大改善推荐的性能。
개성화추천위해결호련망신식과재문제제공료신적사로。위유효지구건용호모형화개진개성화추천적효과，제출료일충알굴비결구화문본중상하문신식적신모형，장득도적상하문신식감입용호모형신식중，봉부료용호모형。실험결과표명，해모형응용우객호대려관평론적상하문수거중，능구대대개선추천적성능。
The incredible growth of information on the Internet is giving more choices but at the same time creating one of the biggest challenges of the Internet,that is,the efficient processing of this growing volume of in-formation.Recently recommender systems have emerged to help users overcome the exponentially growing infor-mation overload problem.In order to form user profiles and improve efficiency of personalized recommendation, an new idea is exploring new data sources such as context information which is one useful data source.This paper has presented a novel approach for mining the contextual information from unstructured text and uses it to produce context-aware recommendations.This method is used to mine hidden contextual information from customers′re-views of hotels dataset and the results indicate that using the contextual information can improve the performance of the recommender system in term of hit ratio.