Now Korea’s crime is showing the tendency of increase and it is on the rise as serious problem. Social interest of crime, prevention and control is piling up because of this. Analyzing the cause of crime is important most of all than others for preventing and controlling the crime.  But in preventing crime, grasping the space structure of social is very important because crime is a social problem. So, this research analyzed SA(Spatial Analysis) and LISA(Local Indicator of Spatial Association) which is spatial autocorrelation that considered factor of space based on the five major crime occurrence data of Seoul from 2011 to 2013. There is many ways that analyzing the cause of crime which consider the spatial factors. It was possible to analyze that grasped the distribution at large, analyzed presence of spatial dependence, found spatial gathering and explored the influence factor of crime etc. The most important one was grasping the presence of spatial dependence. Because if spatial dependence existed, other research could be meaningless. Frist, this research grasped the spatial dependence by spatial autocorrelation. Second, this research grasped the hot-spot, cold-spot and special outlier of research area through LISA. Third, analyzed the flow changes through SA and LISA.

SA can grasped the spatial dependence which could divide to positive spatial autocorrelation and negative spatial autocorrelation. Usually, papers used Moran’s I and Geary’s C for SA. This research used Moran’s I for SA. Moran's I statics defined from -1 to 1. When the statics were lower than 0, they signify negative spatial association, spatial difference. On the contrary, when they were higher than 0, they signify positive spatial association, spatial dependence. Before measure SA, finding the spatial proximity was important. This research used spatial weight matrix for finding adjacency matrix which was the Rook which was chosen because of its normal distribution and Distance which set-up at least that every district had a neighboring areas.

LISA also used Local Moran’s I and the Rook and Distance which is spatial weight matrix for analysis and grasped the four types, HH(High-High), LL(Low-Low), HL(High-Low), LH(Low-High) through the scatter of Moran. HH correspond an area that had a high index and around areas too. LL is opposed to HH. In case of HH, it was an assemblage of high index, so it is called crime hot-spot. The other way, LL is called crime cold- spot. HL and LH is spatial outlier, which had different index between the area and the around area of it.

This Research interpreted the result of LISA with three standard which was group number, types of crime and area. First, group number standard was that the number of hot-spot and cold-spot decreased and spatial dependence decreased either. Second, type of crime was that could find meaningful characteristic in theft, murder and robbery. Theft showed hot-spot mostly gathered in southeast and cold-spot mostly gathered in north. Murder did not show hot-spot or cold-spot and only showed special outlier. Robbery mostly made groups in south. Third, area pattern was that Songpa-gu showed hot-spot and Nowon-gu showed cold-spot for three years. Also, Seocho-gu showed hot-spot and LH and we could know that the crime rate was decreasing. Dongjak-gu and Yangcheon-gu mostly showed LH and this means that the crime rate was decreasing.

Finally, the increase of rape crime was not correlated with the increase of the spatial dependence, so more study to find the exact background is needed. Also, the change from LL type to HL can occur in other crime rates, the background of a murder crime rate change should be known.

This result can use for crime prevention which centering hot-spot area and considering the spatial dependence.