基于案例推理的旅游目的地个性化推荐研究

发布时间：2019-06-05 06:31

【摘要】：随着我国经济和科技的迅猛发展,互联网技术的日益进步,人们生活水平的逐渐提高,人们对能够满足个人兴趣和偏好的自助游的需求也随之增大。由于互联网中充斥着大量信息,用户在网上查找旅游相关信息时难以获得自己需要的有效旅游信息。旅游推荐系统作为解决这一问题的方法成为学者关注的热点,如何向用户推荐能够满足其个性化旅游需求的相关旅游资讯成为研究旅游推荐的关键点。目前的旅游推荐系统通常存在冷启动和数据稀疏的问题,并且推荐内容都是以旅游产品为主,一些为用户推荐旅游目的地的系统也存在推荐的旅游目的地咨询单一,不够丰富的情况。本文着重考虑用户的兴趣,构建以游记为主的案例库,采用基于案例推理的方法构建旅游目的地个性化推荐模型,为用户提供满足其个性化需求且内容丰富的旅游目的地信息,并在一定程度上解决了数据稀疏和冷启动的问题。本文的主要研究工作如下:(1)构建了适合基于案例推理方法的旅游目的地模型和案例用户偏好模型。从"蚂蜂窝"网站中获取用户、游记等信息形成基本案例库,并利用构建的用户旅游偏好算法得到案例用户旅游偏好度。根据用户偏好度和改进的K-Means算法对案例用户按照旅游目的地类型进行分类,形成各种旅游目的地类型子案例库。在检索案例时,只用检索待推荐用户所属类型的旅游目的地类型子案例库,使检索效率得到有效提高。(2)构建了案例属性权重算法,利用调查问卷中用户对旅游要素的评价和案例属性权重算法对案例属性权重进行确定。(3)构建了改进的信任度算法,将信任度引入基于案例推理的旅游目的地个性化推荐系统中,并构建了加入信任度的案例相似度算法,提高推荐结果的准确度。(4)将基于案例推理的方法运用到旅游目的地个性化推荐系统中,利用Mathematica软件对相关算法进行构建和表示,根据用户数据进行实例验证,初步实现了旅游目的地和案例游记的推荐。本文将基于案例推理技术、用户旅游兴趣和信任度的结合,对旅游者实现了个性化的旅游目的地和游记的推荐。既满足了用户的个性化需求,又为用户提供了丰富的旅游信息。通过实例推荐和实证结果,证明了推荐算法的有效性和准确度。本文的研究对个性化旅游目的地推荐系统提供了一定的参考价值。
[Abstract]:With the rapid development of economy and science and technology in our country, with the increasing progress of Internet technology and the gradual improvement of people's living standards, the demand for self-help travel which can meet personal interests and preferences is also increasing. Because the Internet is full of a lot of information, it is difficult for users to obtain the effective tourism information they need when they look up tourism-related information on the Internet. As a method to solve this problem, tourism recommendation system has become the focus of scholars' attention. How to recommend relevant tourism information to users to meet their personalized tourism needs has become the key point of tourism recommendation research. At present, the tourism recommendation system usually has the problems of cold start and sparse data, and the recommendation content is mainly tourism products, and some systems that recommend tourism destination for users also have a single recommended tourism destination consultation. A situation that is not rich enough. This paper focuses on the interest of users, constructs a case base based on travel notes, and constructs a personalized recommendation model of tourism destination based on case-based reasoning, which provides users with rich tourism destination information to meet their personalized needs. To a certain extent, the problems of sparse data and cold start are solved. The main research work of this paper is as follows: (1) the tourism destination model and case user preference model suitable for case-based reasoning are constructed. The basic case base is formed by obtaining the information of users and travel notes from the "polar cell" website, and the tourism preference degree of the case users is obtained by using the constructed user tourism preference algorithm. According to the user preference and the improved K-Means algorithm, the case users are classified according to the type of tourism destination, and the subcase base of various types of tourism destination is formed. In the retrieval of cases, only the tourism destination type subcase base of the type to be recommended is searched, so that the retrieval efficiency can be effectively improved. (2) the case attribute weight algorithm is constructed. The evaluation of tourism elements in the questionnaire and the algorithm of case attribute weight are used to determine the case attribute weight. (3) an improved trust algorithm is constructed. The trust degree is introduced into the personalized recommendation system of tourism destination based on case-based reasoning, and the case similarity algorithm with trust degree is constructed. Improve the accuracy of recommendation results. (4) the method of case-based reasoning is applied to the personalized recommendation system of tourism destination, and the related algorithms are constructed and represented by Mathematica software, and the examples are verified according to the user data. The recommendation of tourist destination and case travel notes has been preliminarily realized. Based on case-based reasoning technology, the combination of users' tourism interest and trust, this paper realizes the recommendation of personalized tourism destination and travel notes for tourists. It not only meets the personalized needs of users, but also provides users with rich tourism information. The effectiveness and accuracy of the recommendation algorithm are proved by example recommendation and empirical results. The research in this paper provides a certain reference value for personalized tourism destination recommendation system.
【学位授予单位】：海南大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F592

【参考文献】