Research On The Evaluation Method Of Livelihood In Urban Residential Area Based On WEB Text Information Mining

With the rapid development of China's economy,people's living environment requirements are getting higher and higher, from the original "home to seek housing"changing into the "home to select housing" . Livable living environment has attracted people's more and more attention. In the purchase or rental, the residential area is a livable has become an important reference for people's decision-making. At the same time, the state has also introduced a number of measures to promote the creation of livable living environment, China will be selected "Chinas top ten livable city."every year since 2007.In the whole people to participate in the creation of livable life environment, livab le evaluation standards are not perfect and related data acquisition difficulties, affected the livable living environment construction seriously . In addition, with the continuous deepening of the degree of national information, real estate network platform for the rapid rise of this study provides a massive amount of data.Through the Web text information acquisition and mining,This paper establishes the evaluation model of urban residential area livability,which is based on Web text information mining,and then evaluates the livability of urban residential areas,finally takes the central area of Wuhan as an example to evaluate the livability of residential area to prove the feasibility of the proposed research method,as well as to help for the further study of urban planning and livable community construction.The main contributions and innovations of this paper are as follows:1. A targeted Web information mining method is given in this paper. In order to quantify the objective evaluation index data (commercial facilities, educational facilities,etc.) of residential area, a more scientific and practical method of quantification is given by using the characteristic word statistics and the actual situation and related standards. In this paper, an improved KNN classification algorithm based on the local sensitive hash method to reduce the computational complexity and improve the efficiency of text classification is proposed in order to improving the classification efficiency of KNN classification algorithm in text classification of subjective evaluation index data. In this experiment, the improved classification algorithm has a significant improvement in the classification rate compared with the original classification algorithm in the case that the classification effect is basically the same.2. The evaluation model of livability of urban residents based on Web text information is established. The data required for the evaluation model are from the Internet,therefore,in determining the model evaluation index is based on the existing theory of livability and combined with the characteristics of the relevant Internet data can be obtained.At last,the evaluation index is divided into objective evaluation indicators (such as the evaluation of the indicators, Residential area hardware and software data) and subjective evaluation indicators (that is, residents of the residential area of the subjective evaluation) two categories. The evaluation results are compared with the actual situation,and the feasibility of the evaluation model operation and the accuracy of the evaluation results are proved by comparing the evaluation results with the actual situation in the central area of Wuhan.3.A improved comprehensive evaluation method of the evaluation model of livability of urban residents is posed. Aiming at the situation of subjective and objective combination of urban residents' livability evaluation model, a principal component evaluation method based on subjective and objective weights is proposed. Compared with the original principal component evaluation method, this method not only retains the advantage of objectivity, that is, the proportion of each objective index data is determined by its own relevance, and the subjective factors are taken into account. Important influence, that can manually adjust the subjective and objective indicators of the weight coefficient.Experiment shows that the evaluation results of the improved method are more in line with the actual situation.