基于多源数据的选址模型及其应用研究

发布时间：2018-05-25 05:27

本文选题：智慧城市 + 多源数据集　；参考：《电子科技大学》2015年硕士论文

【摘要】：随着智慧城市和大数据技术的发展,多源数据深刻地改变了商业行为。其中,对商业店铺的选址影响也尤为明显。由于消费者对商家的需求逐渐上升为对购物便利性的需求,为更好的服务大众,商家在改善经营技术外,主要依赖于商业店铺的选址技术。针对商业店铺的选址问题,目前使用较多的方法是基于专家咨询的层次分析法,这种方法通过把总是和决策相关联的元素划分为不同的层次,凭借专家的个人经验和专业知识对不同的元素进行重要性打分,最终得出相关决策。由于该方法需要利用专家打分对各个指标因子进行评价,专家的个人主观因素较强。因此,如何设计一种无需引入人为因素的方法显得尤为重要。本文的主要工作可以概括为以下几点:(1)提出了一种基于多源数据集的选址模型建立方法。结合相关数据集,从中选择出一些可能对选址产生影响的特征指标,使用支持向量机(Support Vector Machine)技术对这些特征指标进行学习,得出选址模型,然后使用该模型确定在候选区域中哪些位置选址可能使商业店铺有较好的发展前景。(2)根据选址建模方法的功能需求,设计了一个基于支持向量机技术的选址建模系统并对各个功能模块进行了具体实现,然后通过一个实际的应用案例对上述选址方法进行了测试评估。测试的结果表明,使用支持向量机得出的选址模型具有较高的准确性。由此证明了基于多源数据集建立选址模型这一方法的可行性和有效性。理论和实际应用表明,本文提出的基于多源数据集的选址模型建立方法可以用于解决商业店铺的选址问题,该方法能够节省人力物力,具有一定的实用意义和借鉴价值。与此同时,该方法也为选址问题的研究提出了新的思路。
[Abstract]:With the development of smart city and big data technology, multi-source data has profoundly changed business behavior. Among them, the location of commercial stores impact is particularly obvious. Because the demand of the consumer to the merchant gradually rises to the demand for the convenience of shopping, in order to serve the public better, the business mainly depends on the location technology of the commercial store besides improving the management technology. In order to solve the problem of business store location, the most popular method is the analytic hierarchy process (AHP) based on expert consultation, which divides the elements that are always associated with the decision into different levels. Depending on the expert's personal experience and professional knowledge to evaluate the importance of different elements, and finally draw relevant decisions. Because this method needs to use the expert score to evaluate each index factor, the expert's individual subjective factor is stronger. Therefore, how to design a method without introducing human factors is particularly important. The main work of this paper can be summarized as follows: 1) A method of setting up a location model based on multi-source data sets is proposed. Combining with the relevant data sets, we select some characteristic indicators that may have an impact on location selection, and use support Vector Machine (SVM) technology to study these feature indicators, and obtain the location model. The model is then used to determine which locations in the candidate areas may provide a good prospect for commercial stores.) according to the functional requirements of the location modeling method, A location modeling system based on support vector machine (SVM) technology is designed and implemented for each functional module. Then, a practical application case is used to test and evaluate the location selection method mentioned above. The test results show that the location model based on support vector machine is accurate. The feasibility and effectiveness of the method based on multi-source data set are proved. The theoretical and practical application shows that the method proposed in this paper can be used to solve the location problem of commercial stores, which can save manpower and material resources, and has certain practical significance and reference value. At the same time, this method also provides a new idea for the study of location problem.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F717;TP181

【相似文献】