摘 要
推荐系统旨在为用户推荐智能化的在线商品或信息，其广泛应用于众多Web场景之中，来处理海量信息数据所导致的信息过载问题，以此提升用户体验。鉴于推荐系统强大的实用性，自20世纪90年代中期以来，研究者针对其方法与应用两方面，进行了大量广泛的研究.近年来，很多工作发现知识图谱中所蕴含的丰富信息可以有效地解决推荐系统中存在的一系列关键问题，例如数据稀疏、冷启动、推荐多样性等。
近几年随着数字图书馆的高速发展，其数字资源馆藏已实现了海量增长，读者对资源的需求越来越智能化，这些因素推动着图书馆推荐系统应用的普及和深化。知识图谱技术是现今图书馆学的研究热点，据此，对基于知识图谱的图书馆推荐系统进行了研究，并将其与传统的基于ALS的协同过滤算法进行比较.结果显示，基于知识图谱的MKR推荐算法可以取得较好的图书推荐效果，能够助力今后智能化图书馆推荐系统的建设。
本次图书智能推荐系统前端采用Bootstrap框架，后台采用Java的SSM框架和Mysql数据库，将系统分为了三层：Web层、服务与模块层、数据层，并对每层的结构与需要完成的功能做了定义。并建立了图书智能推荐系统所需的数据库，该数据库的数据共分为三个部分：原始数据、清洗后数据和用户数据。设计了图书智能推荐系统的页面，主要包括首页、搜索、展示、登录、注册等页面。经过后期的数据库优化与功能测试，系统与同类网站相比，性能良好。
关键词：知识图谱；个性化；智能化；图书推荐；SSM
Abstract
A recommendation system is designed to recommend intelligent online goods or information for users. It is widely used in many Web scenarios to deal with the information overload caused by massive information data so as to enhance the user experience. In view of the strong practicability of the recommendation system, since the mid-1990s, researchers have carried out a lot of extensive research on its methods and applications. in recent years, many works have found that the rich information contained in the knowledge map can effectively solve a series of key problems existing in the recommendation system, such as data sparsity, cold start, recommendation diversity and so on.
In recent years, with the rapid development of digital library, its digital resource collection has achieved a massive growth, readers' demand for resources is becoming more and more intelligent, these factors promote the popularization and deepening of the application of library recommendation system. According to the research focus of library science, the knowledge map-based library recommendation system is studied and compared with the traditional ALS-based collaborative filtering algorithm. Results show that the MKR recommendation algorithm based on knowledge map can achieve better book recommendation effect and help the construction of intelligent library recommendation system in the future.
The front end of this book intelligent recommendation system adopts Bootstrap framework, and the background adopts Java SSM framework and Mysql database. The system is divided into three layers: Web layer, service and module layer, data layer, and the structure of each layer and the functions to be completed are defined. and the database needed for the book intelligent recommendation system is established. the data of this database is divided into three parts: raw data, data after cleaning and user data. Design the book intelligent recommendation system page, mainly including home page, search, display, login, registration and other pages. After the later database optimization and function test, the system has good performance compared with similar websites.
Keywords: knowledge map; individuation; intelligence; book recommendation; SSM
