基于关联规则和决策树对大众数据的商业选址的分析与应用报告方案.doc

课 程 论 文 （ 2015/ 2016 学年 第一学期） 课程名称 数据挖掘 论文名称 写作时间 年 月 (系) 管理学院 专 业 信息管理 与信息系统 基于大众数据的商业选址 摘要 在经济快速发展和市场竞争愈演愈烈的同时，不知不觉我们也已进入和生活在数据时代，我国各种产业都进入了一个更为激烈的竞争环境，尤以商业和房地产业为主。而目前我国许多商业机构普遍存在着各种问题，比如效益低下，地理位置不佳、规划性不够等问题。为了解决企业所面临的一些问题，本文主要研究了基于大众数据的商业选址问题以及选址问题背后所隐藏的发现，同时以国内非常普及和使用的网站“大众点评网”作为研究问题的主要数据来源，虽然数据挖掘方法众多，但是主要用了关联规则和分类的方法来进行深入剖析，并利用非常流行小巧且开源的数据挖掘软件Weka做以辅助分析，主要分析地理位置对公司企业的是否会影响，地理位置对商业机构具有怎样的重要性，如何正确选择地址对商业机构的效益是否有增加，同时展示图表以增加计算结果的可视化，在将显示结果和生活常识的结合下，预测商业选址对房地产产业开发是否具有一定的意义和参考价值。 关键词：选址 关联规则 决策树 The commercial location based on mass data Abstract In the rapid development of economy and the growing competition in the market at the same time, we also have entered imperceptibly and living in the age of data, a variety of industry of our country has entered a more fierce competitive environment, especially in the commercial and real estate. While many of the current commercial institutions in China generally exist various problems, such as low efficiency, poor location, planning is not enough etc.. In order to solve some of the problems faced by the enterprises, this paper mainly studies the mass data of commercial location problem and the location problem of hidden behind the discovery based on, at the same time to home is very popular and the use of the site the masses comments on a net as the main data source of research problems, although the number of data mining methods, but the main use of the method of correlation analysis and the classification and prediction of statistical method, to carries on the thorough analysis, and using the very popular small and open source data mining software Weka to do to assist the analysis, the main analysis of the geographic location of the company would affect, geographic location is how important to businesses, how to choose the correct address to the commercial institutions benefit if there is an increase in at the same time chart, to increase the visualization of calculation results, combined with the results