In the field of tourism,tourists often get the information they need on the Internet before traveling,but the phenomenon of in-formation overload online in tourism industry is becoming more and more serious,so that personalized information cannot be obtained by users. The problems of sparsity and scalability exist in the traditional tourism recommendation algorithm based on collaborative filtering, and sometimes users can’ t express their needs and can’ t be satisfied with the recommendation based on the knowledge of the recommen-dations. For these problems,a tourism recommendation algorithm based on fuzzy clustering is proposed,which is used for the users to rec-ommend the tourism products that meet their needs and preferences. Tags are used by the algorithm to build user’ s preference models and sights feature attribute model,fuzzy clustering on them. A new similarity measure is proposed. On this basis,the combination of content-based and collaborative filtering technology is recommended. Experimental results show that the proposed algorithm can significantly im-prove the efficiency,scalability and accuracy of the recommendation system.%在旅游领域中，旅游者常常在旅游前从互联网上获取所需信息，但是在线旅游业日益严重的信息过载现象，使得用户不能得到他们想要的个性化信息。传统的基于协同过滤的旅游推荐研究普遍都存在稀疏性和可扩展性等问题，基于知识的推荐研究有时因用户无法表达清楚他们的需求而无法得到满意的推荐。针对已有的旅游推荐算法存在的问题，提出了一种基于模糊聚类的旅游推荐算法，为用户推荐符合其需求和偏好的旅游产品。该算法利用标签构建用户偏好景点模型和景点特征属性模型，对数据集进行模糊聚类，同时提出新的相似度度量。在此基础上，组合基于内容和协同过滤技术进行混合推荐。实验结果表明，该算法能显著提高推荐系统的效率以及可扩展性和准确度。
展开▼
