基于用户历史瓦片浏览记录的兴趣点智能搜索方法优化研究

Optimization of Intelligent Search Methods of Point of Interest Based on User Browsing Records

在网上各种信息呈现爆炸式增长的背景下,传统方法往往难以顾及用户兴趣,而导致查询无法体现个性化。针对这一问题提出一种基于用户历史瓦片浏览记录的兴趣点智能搜索方法。首先对智慧城市平台下的用户历史瓦片浏览记录数据进行分析,并以热力图可视化形式展现出用户关注的热点区域,得出空间热度;然后根据属性查询中加入空间热度影响因子来影响兴趣点搜索结果,使搜索结果更加符合用户搜索意图,针对热点区域给不同用户以相应服务;以滕州市兴趣点为实验数据,使用Elasticsearch构建索引数据库,采用Web前端技术搭建搜索框架,经过实验对比分析发现,该方法能够有效地提高搜索的查准率,并且可随着数据量的丰富变得更加智能化。

The Internet information grows explosively today.However,traditional search methods usually ignore users’interests,which lead to non-personalized search services.To solve this problem,an optimized intelligent search method of point of interest is proposed.Firstly,the user browsing records of the smart city platform are analyzed and a heat map is drawn to visualize the hotspots that attract great attention of users and to calculate the space heat.Then,the space heat has been added to the attribute query to obtain search results which are more consistent to users’search intentions.Corresponding services can be provided to users located in different hotspots.Finally,Taking POI of Tengzhou city as an example,this paper uses Elasticsearch to built an index database and develop a search framework based on of Web front-end technology.Results indicate that the proposed approach can significantly improve search accuracy and is more intelligent when data volume increases.