Key Technologies Research In JavaEE-based Forest Land Management WebGIS

With the growing contradiction between the exploitation of the forest and the environment, how to effectively manage forest resources to maintain the forestry sustainable development, has become a common concern of the international community. Geographic Information System(GIS) is a platform of geographic data collection, storage and management for effective decision-making and planning of the sustainable forestry development.Combining Web technology and GIS technology, Web Geographic Information System (WebGIS) is the expansion of traditional GIS,so it has a more extensive application prospect in future.Geographic data is the resource base of GIS, its access and update is a prerequisite to run GIS. Geographic data contains spatial data and non-spatial data,having the character of high dimension and locality. With the development of database technology and remote sensing, forestry database system accumulates more and more spatial data. How to extract potentially useful information from vast amounts of geospatial data becomes the great challenge of WebGIS. Datamining is the process to find the unknown and the value of the model from a large number of data. This thesis combines the WebGIS with the datamining techniques,using four browser/server mode, and researchs the key technologies in JavaEE-based forest land management WebGIS.On the basis of previous study results, With deep research to data mining techniques in the rules relevant algorithm,this paper proposes a new algorithm (TIMBM-Apriori)。As woodlands webgis support module,TIMBM-Apriori is the important implementatal technology.The main contents of this paper can be summarized as:(1)In the system's implementation,to solve the problem of the high dimension and continuous attributes effect,this paper introducts K-means clustering method for data pre-processing. (2)This paper uses the classic association rules algorithm Apriori as the primary analytical tool.By using Apriori algorithm,a large number of data set will result in combinatorial explosion problem. And in scanning large amounts of data, the algorithm is obvious inefficiencies.To solve these problems, this thesis proposes a Transaction-Item Mapping Boolean Matrix(TIMBM) Apriori algorithm, the algorithm only need to scan transaction data once, and then convert the transaction database to Transaction Bool Matrix(TBM), which can greatly reduce database I/O operation time, and the cost of increasing the speed just open up a certain amount of memory space to store TBM. The experiment with Yongan sub-compartment map data in Fujian shows the better application advantage than the traditional Apriori algorithm.TIMBM-Apriori algorithm has more efficiency and better quality of mining results, especially for time-sensitive WebGIS.(3)This paper proposes a great solution based on the analysing the main problems of the current forest management system. Through deeply analysing specific demand, designing the system's functional structure, and implement the following function:basic GIS function,Data integrated query function and decision support function based on TIMBM-Apriori algorithm. All of these are the key functions in this system.