Study On Development Level Of Urban Space Based On Open Data From Website

With the rapid growth of ICT(information communication and technology)usage,elements of region or urban development start to accelerately flow to receive a new space called "space of flows",which will become the main carrier for activities of governments,enterprises and residents.The traditional urban space study based on small data should be transformed to meet the change of spatial organization.Meantime,the ICT also timely records activity data(big data)of governments,enterprises and residents,which can be used to resolve the difficult problems in urban space study instead of usage of traditional small data.For instance,simulation of resident activity in a large scale.Urban researchers start to questioned the scientific of small data and cheer the coming of big data era.But,this upsurge is easy to turn into "big data determinism”to overlook the role of small data.In recent year,Chinese government make effort to new urbanization development and smart city construction in order to better meet resident demands and improve their life essentially,which is keeping with the people-oriented in urban space study effected by ICT.Based on above new tendency,this thesis finds that there are no complete methodology and method system for urban space study in the information era.How to combine with the small data when we use big data to research urban space?And how to help the specific empirical study by using the new method system?These are also the research targets of this thesis.According to the summary of related urban space studies,this thesis firstly puts forward to the necessity and direction of method framework for urban space study from methodology perspective in the information era.Secondly,based on the analyses of big data in the type and processing method,the thesis gives the comprehensive research framework of urban space in the information era,and also builds two detailed method frameworks,namely urban space effect and urban space development in the information era.Thirdly,it focuses on the method framework of urban space development(more precisely,development level and factors of urban space),and takes restaurants of Nanjing city as a case to verify above method framework.1)It collects open big data from dianping website to analyze the spatial characteristics and structure of restaurant popularity in Nanjing city by using principal component analysis,cluster analysis,and Kernel density tool,and also gives some advises for space optimization;2)Combined with scores of restaurant popularity,this thesis uses multiple linear regression to find the factors of restaurant popularity in Nanjing city,and also put forward to several suggestions for restaurant owners in management options and store’s address choosing.The main conclusions are as follows:As to the methodology,this thesis believes that urban space study in the big data area is not only such issues in the data or technology level,but also the thinking transformation of traditional urban space study.It concludes three combinations in urban space study in the information era,which are as follows:1)Physical space and Activity space.Urban space is composed of physical space and activity space,both of which affect and supplement each other.The physical space is a carrier and also affects resident activity.The resident activity changes the specific structure of activity space timely and promotes the optimization of urban physical space.2)Correlativity and Causality.The former mainly based on big data often answers to the "what",while the latter mainly based on small data often refers to "why".The urban space is a complex place for all types of elements,a large number of information,and all kinds of activities that it is necessary to understand the relationships among all systems within the space and their changing mechanisms.3)Macro-analysis and Micro-exploration.It is not only needed to Multi-angle and Multi-type analyze the urban space in a large scale,but also pay more attention to individual behavior and preference.Therefore,comprehensive research in all scales is a inexorable trend by combining with small data and big data in the future urban studies.As to the method framework,with the rapid development of people-oriented new urbanization and the appearance of resident activity data recorded by ICT,the traditional urban space study will change its research focus that the relationship between urban space and resident activity(or behavior)will become the major field for urban space study in the information era,which is also the important application for combination between big data and small data.In particular,taking the urban space as research object,it is needed to discuss the impacts of resident activity on urban space,and vice versa.This thesis put forward to three method frameworks,namely AADSF("microscopic Analysis of Activities+Delineation of activity Space+Factors discussion"),MFVE("Model building for small samples+Factors discussion+Verification and Exploration for big data"),and ECDF("spatial Evaluation for big samples+spatial Characteristics and Differences discovery+Factors discussion for small samples").Among them,the AADSF and the MFVE can be used to research the urban space effect in the information era,and the ECDF is benefit to the study of urban space development in the information era.As to empirical study,this thesis finds two conclusions based on analysis to prove the method framework of urban space development,and also give advises for space optimization and restaurant owners’ management or decision in Nanjing main city.1)Development level of catering space in Nanjing main city:the distribution of high-level restaurants shows a tendency of "Two cores and Multiple centers",and the number and scale of high-level restaurant centers are both less than the ones of restaurant density centers;different types of restaurants have different distribution of development levels with all restaurants;the hierarchical structure of restaurant development level in Nanjing main city is not consistent with the principle of K=3 from Central Place Theory,which may be related with urban spatial structure,customers’behavior,and online information search.2)Factors of restaurant development level in Nanjing city:the significant factors includes restaurant types,restaurant size,management model of restaurant,number of restaurant branches,taste score of dishes,atmosphere of restaurant,the highest basic land price of business center,number of comparable restaurant 300 meters in the range;However,the per capita consumption,special service score,road grades,location,and population density of neighborhood have no significant effect on restaurant development level in Nanjing city.Generally,this thesis preliminary enrich the previous studies to put forward to the methodology of urban space study in the information era,which is much more helpful to research selection and method combination of urban space study in the information era.Meantime,the three detailed method frameworks also improve the level of traditional urban space study to give more advises for data collection,method selection and result application in urban space study,especially in urban China with development of new urbanization and smart city construction.Moreover,this thesis uses big data from online comments and combines with statistical and spatial data to comprehensive research the development level of restaurants in Nanjing city,which can reduce the influence of limitation for lack of development level data of traditional studies.