Clustering Analysis Of Temperature Data In Liaoning Province

With the continuous development of technology,all areas are facing a major problem of data that can not be avoided,the traditional method of face large GIS data processing and continue to highlight the disadvantages of the application on the geographical data in temperature and precipitation data over constantly updated every day accumulate,face a lot of ground stations continue to provide the data,how to use efficient methods and algorithms utilized,our geography people to keep pace with the times to think,to find breakthrough algorithms to use and Geography the combination is an urgent need to do the same on the interpretation of remote sensing geographical information systems also face the problem of big data,with the continuous improvement of remote sensing technology,high spectral image generation,remote sensing interpretation to also rely mainly on manual operation way we challenge the workload on the face of these problems this paper,data mining and geography combine to explore how to quickly and effectively solve the problem of geographical data.Liaoning Province,the Bohai economic zone as a ring,making it the revitalization of northeast old industrial base of the task has a pivotal position,how to use the law of geography methods to study the geographical and meteorological factors in the region to the spatial and temporal distribution "old industrial base," the economic rise and provide a reliable theoretical method is an urgent problem.Data mining algorithm can process large data problems,we rely on dimensionality reduction and clustering analysis of temperature of 23 meteorological stations in Liaoning Province were studied on the spatial and temporal distribution,obtained Spatial distribution of temperature,and compare it with the traditional clustering methods,the new LLE-FCM algorithm model used herein,than simply clustering algorithm to be more suitable for the actual operating data,the innovation is mainly reflected in two points:First,clustering algorithm better,more reasonable highlights the spatial and temporal distribution of temperature in Liaoning province;the second is to reduce the number of runs of the algorithm to improve the processing speed.Realization of the above two points make data mining solutions to the problem of geographical data in GIS and a small step forward.