Study Of Data Mining And Visualization Based On Yellow River Delta Marine Geographical Information Data Warehouse

Data mining, known as Knowledge Discovery in Databases (KDD), is defined as the nontrivial extraction of implicit, previously unknown, and potentially useful information from a great volume of data accumulations. It uses machine learning, statistical and visualization techniques to discover and present knowledge in a form being easily comprehensible to humans. Data mining is an advanced data processing technique. With the rapid development of survey and information technology, our skill of collecting, storing, and transferring Yellow River Delta marine geographical information data has been improved dramatically. A great of rich information that we have not discovered before exists in the YRD data warehouse. In order to take full advantage of it, we have designed and developed a scientific and applicable data miner system, Yellow River Delta Data Miner (YRDDM).The frame of this paper, based on the Yellow River Delta marine geographical information data flow, is organized as followings. 1) data management, 2) data mining, and 3) visualization. In order to manage the data efficiently, the metadata database is firstly constructed according to YRD database, and data warehouse is set up based on large volume of YRD data. Secondly, many kinds of algorithms of data warehouse have been designed and realized in our YRDDM. Lastly, in order to understand more spatial information with an intuitionistic way, visualization technology has been applied to display the analysis results. YRDDM, which aims at the management, analysis and visualization of Yellow River Delta marine geographical information data, is a seamless integration of these three parts.YRDDM is applied into the YRD scientific research as soon as it has been funded. The BP neural network is used to restore the underwater landform of Yellow River Delta. The SOM neural network is used to classify the RS images'information. The Regression Model is employed to analyze the section of water depth. The Grey Model is used to forecast the area Yellow River Delta. The Cluster Model is used to cluster the polluted station of Bo Hai. The PCA Model is used to estimate the polluted level of Bo Hai. The SVD Model is used to analyze the salinity, temperature and concentration data which were sampled from the estuarine of Yellow River in "973" program. Furthermore, we illustrate the data management and visualization model particularly by use of corresponding examples.Characteristics and shortcomings of YRDDM have been summarized at the end of our work, pointing out the further developing orientation and its potential prospect.