The Research On Address Matching And Its Application In Urban Disease Mapping And Spatial Analysis

In recent years, with the continuous development of computer technology, geographic information systems and spatial analysis, such as these technologies provide a good fundational platform for the further analysis of a large number of accumulated disease data and explore its space-time trends for the public health department. GIS is a powerful tool for spatial data processing, information visualization, spatial statistics analysis and diseases mapping which offers a new perspective and point of view for urban epidemic.This article first analyzed the advantages and disadvantages of address model from international comparison, according to the address coding standards in our country. This paper put forward a kind of adaptive address model based on association rules, and adopting the model for shenzhen address data modeling. On the basis of address elements, combined with the key words in address elements as a supplementary means to explore the association rules between address elements, using the associated high frequency address elements to design a unified address model in shenzhen. Secondly, the paper introduced address matching principle in detail, and designed the address database according to the address of adaptive model based on association rules. This paper determines the databases storage process based on different type of address and inspects the address data quality. Then we put forward a kind of reverse alignment based on address elements levenshtein algorithm in address matching, the method optimizes the levenshtein address matching algorithm and speed up the address matching process and improve the efficiency of address matching. And this method achieved a good effect for the spatialization of shenzhen liver diseases data during 2010 to 2012. The address matching technology provides a good implementation ideas and methods for the spatialization of urban disease data. Lastly, the spatialization of urban disease data based on address matching technology was used for disease mapping and spatial analysis. We analysis the spatial-temporal distribution of Shenzhen liver diseases data, and use the spatial-temporal scan statistics method to explore the space-time aggregation, including aggregation areas and aggregation time. Then the relative risk analysis is used for the decision of the government department of health and the social public behavior, and the work provides technique for disease prevention and control.This study takes shenzhen city as the study area and the diseases data come from liver disease hospital registration data in 2010-2012, through address matching and spatial analysis to explore the disease mapping and spatial statistic analysis of urban epidemic, the research can be used to prevent diseases for the social public and reasonable allocation of medical resources for medical institutions and provide a scientific basis for health management departments to make decisions.The main research contents are as follows:(1) The paper introduce the main theories and methods for city disease data address matching and mapping spatial analysis, including the definition of disease mapping research, spatial effect of geographical disease data and disease mapping between population distribution and spatial-temporal distribution. Then spatial analysis method for urban epidemic data was introduced in detail, such as spatial statistical analysis, exploratory spatial data analysis, spatial-temporal clustering analysis, relative risk analysis, multi-scale clustering characteristics analysis, also elaborated the advantages in the spatial data analysis method for the analysis of urban disease mapping, finally introduces the common spatial statistical analysis softwares.(2) The paper elaborates the address model researches at home and abroad. On the basis of address elements, we count stern word frequency in address elements and calculate support and confidence of address elements combination to explore the association rules between address elements and combination model, this paper proposes a adaptive address model based on association rules method for address information sharing and address matching service to provide suitable model.(3) This article adopts addresses matching levenshtein algorithm based on reverse aligned address elements for shenzhen diseases data spatialization. We use shenzhen liver disease hospitalization data from 2010 to 2012 to verify the address matching effect, the matching rate of the liver in hospital registration data is as high as 95% which demonstrated the improved levenshtein algorithm speed up the address matching and improve the address matching precision.(4) The paper adopts the spatialization Shenzhen liver data for disease mapping and spatial analysis. Spatial statistical analysis method is used to describe the "three" distribution of shenzhen liver data and visualization. We explore the spatial-temporal aggregation of liver disease in shenzhen city by spatial-temporal scan statistics measurement and from multi-scale perspective such as the macro, meso and micro analysis the space-time distribution pattern and trends, then analysis the relative risk and the migration of the gravity of shenzhen liver disease to provide a scientific basis for early prevention of liver disease. We analysis the advantage of the spatial statistics method in disease mapping, and through the clustering characteristics of multi-scale analysis further states the scale effect of disease mapping and spatial heterogeneity. The space-time aggregation test and relative risk assessment provides a data basis for the cause of disease and a reference basis for further disease prevention can be controlled as early as possible.In conclusion, the paper discusses the problems encountered in the process of this study, and summarizes the research work and the framework and put forward the the research direction in the future.