改进TOPSIS法用于出租房源综合评价研究

COMPREHENSIVE EVALUATION OF RENTAL HOUSING SOURCE BASED ON IMPROVED TOPSIS

综合目标地点和出租房源的面积、租金及间隔目标地点的距离等属性因素对出租房源进行综合排序研究。这对侧重点不同的租房者进行出租房源选择时具有较高的实用价值。通过对传统TOPSIS法缺陷的研究,提出一个基于最差理想解和马氏距离的改进TOPSIS法,用于对出租房源进行综合评价。该改进方法通过相关系数改进马氏距离,利用改进后的马氏距离替代欧式距离,解决传统TOPSIS法因出租房源的面积、租金等属性间相关性导致欧式距离失效的问题;根据传统TOPSIS法的正负理想解提取一个最差理想解,并用其代替一般TOPSIS法的负理想解,解决出租房源间隔正负理想解等同距离时无法准确确定位置的问题;通过租房者偏好系数将马氏距离和最差理想解所得距离尺度合成新的相对贴近度,对出租房源进行综合评价。实验证明利用该改进后的TOPSIS法排序的出租房源的推荐结果更加合理。

The comprehensive ranking of rental housing sources was studied by combining the attribute factors of target location and the area of rental housing, rent and distance between target locations. It is of great practical value to select rental housing sources for different renters. In this paper, an improved TOPSIS based on the worst ideal solution and Markov distance was proposed by studying the defects of the traditional TOPSIS, which could be used for comprehensive evaluation of rental housing resources. The method improved the Markov distance by correlation coefficient, and then replaced the Euclidean distance with the improved Markov distance. It solved the problem of Euclidean distance invalidation caused by the correlation between the area and rent of the rental housing resource in the traditional TOPSIS. A worst ideal solution was extracted from the positive and negative ideal solution of traditional TOPSIS. It replaced the negative ideal solution of the general TOPSIS, and solved the problem that the positive and negative ideal solution of the rental housing source spacing cannot accurately determine the location when the distance was equal. Through the renter preference coefficient, the distance scale obtained from the Markov distance and the worst ideal solution was combined into a new relative closeness degree, and the rental housing resources were evaluated comprehensively. Experiments show that the recommended results of the improved TOPSIS are more reasonable.