The application of data mining in multi-supplier Points of Interest processing

With Digital Earth´s gradual deepening into people´s life, electronic geographic data are employed in more and more industries, including the use of position information and service information, as well as the application of the digital road and the digital POI (Point of Interest). A powerful and rich dataset of digital POI is expected. However, digital map products from different enterprises have their own characteristics and are quite different from each other. Integrating and unifying these products can not only save cost but also provide a more comprehensive and complete digital map product for users. This paper is aimed at establishing a comprehensive processing framework for multi-vendors´ POI data. Similarity metrics and data mining algorithms are combined to predict and classify the multi-attribute similarity measurement results. Experimental comparison and application results verify that the proposed methods can not only solve the problems of business processes being independent, and the processing framework as well as hardware resources being unshared, but also improve efficiency and reduce labor costs for data processing.