Research On The Identification And Optimization Of Public Space Vitality And Its Influencing Factors Based On Machine Learning-GIS

As an indispensable space type in the city,public space plays an important role in people’s work and life.At present,the public space of most cities in our country still needs to improve its vitality,but the traditional public space planning is mainly based on macro strategies,lacking of targeted optimization programs.In recent years,with the significant improvement of computer computing power,combining massive network data with machine learning,spatial statistics and other related algorithms,we can study the vitality of urban public space more precisely and timely,and then put forward targeted optimization measures.In this paper,236 public spaces in the central urban area of Tianjin are selected as the research object.Based on massive network data,the vitality of public space and its influencing factors are quantified by means of machine learning image semantic segmentation algorithm(SEG net)and spatial statistics,and the spatial distribution characteristics of the two spaces are determined.Then identify the optimization area of public space vitality,mining the influencing factors of the main types of public space vitality.Finally,it puts forward targeted strategies to reshape the vitality of public space.The first part analyzes the spatial distribution characteristics of the public space vitality and its influencing factors in the central urban area of Tianjin,which provides the basis for identifying the optimized area of the public space vitality later.In this part,first of all,by summarizing the existing research results,we propose to use the relevant data to quantify the vitality of public space and its influencing factors with the help of the combination of machine learning image semantic segmentation(SEG net)and spatial statistics.Secondly,using the classical statistical method to test the difference of public space vitality between working days and rest days in the study area,and identify its spatial distribution characteristics by hot spot recognition,high-low clustering and other methods.Finally,the same method is used to identify the spatial distribution characteristics of 7 categories and 16 influencing factors of public space vitality.The second part,based on the results of the previous analysis,identifies the areas and main types of public space vitality in the research area,and excavates the influencing factors of the main types of public space vitality,which can provide quantitative basis for the follow-up optimization exploration.Firstly,this part summarizes the spatial distribution characteristics of public space vitality and its influencing factors(high and low value concentration area),and identifies that the optimization area of public space vitality mainly includes 7 areas,such as the north side of Tianjin station.Secondly,by comparing the number and proportion of each type of public space in the region,the author thinks that the vitality of three types of public space should be optimized: Community Point Park,community point green space and district level point park.Finally,through the combination of geographical detector and linear regression,the influencing factors of the above three types of public space vitality are explored.The third part puts forward the optimization exploration of public space vitality and its influencing factors in the research area.First,according to the results of the second part of the analysis,this paper puts forward three types of public space vitality optimization strategies: Community Point Park,community point green space and district level point park.Secondly,in the empirical exploration,the preliminary analysis is integrated,taking the representative public space of the point green space type in the north of Tianjin station as an example,and the corresponding optimization scheme is proposed.The author hopes that the results of this study can provide a reference for the planning and construction of public space in the central urban area of Tianjin,and this research process can provide constructive suggestions for the construction of public space in other cities.