A thesis submitted in partial fulfillment of the requirements for the degree of Master Science in Remote Sensing and Geographic Information Systems

Thesis (M.Sc.) - Asian Institute of Technology, 2009

Geocoding is the process of assigning geographic coordinates (often expressed as latitude and longitude) to locations (street addresses, zip codes). Geocoding is becoming increasingly important in many application areas that involve the analysis and mining of data. Some of the more important disciplines that rely on accurate geocoding for spatial analysis include crime mapping at law enforcement institutions, health database geocoding and linking at health institutions, commercial customer database geocoding, environmental applications using point source data, and a multitude of municipal government applications. Using geocoding for Thailand will have many advantages in both public and private sectors. The important algorithm for geocoding is addressed interpolation/address matching. Address matching has been used to identify geographical coordinates from addresses. Moreover, the most important information required for geocoding is reference data/street centerline data. Address range, its coordinate and street name are the most important information. The problem of existing technique which can not be applied for Thai address is an insufficient of the street centerline database in Thailand. It does not have enough information required for geocoding/ interpolation such as address range. In this research, author compare the information of street centerline data between the United States and Thailand, and identifies problems of apply existing technique to Thailand. However, author proposes new solution to use addresses of POls (points of interest e.g. school, university, company, temple, etc.) for interpolating and extrapolating the address/house number. In this research, author proposes new method of geocoding for Thai address by using linear method. In linear method, it is used the address of POls to calculate location of input address. In conclusion, on the result linear method, author found that this method can be applied if it has enough POls (Points of interest) with address. The accuracy depends on number of POls and distance between POls. However, for better results, it should be applied to chronological Thai address system