Technology On Fast Visualization Of Large Three Dimension Scatter Plot Based On Datacube

In recent years, data mining technology has got more and more attention for its using in knowledge discovery. In order to make the interaction processing with users more easily, data visualization has a very important application in data mining process. Therefore the development of data view technology is also very important. However, with the increasing scale of datasets, the data visualization has poor performance in term of speed. It makes users wait too long to lose patience and it also lacks of many interaction operations.For solving these problems, this paper has designed a data cube visualization model to store and compute the data items in a more quickly way. Based on the distribution of the visualization datasets, the sample dataset of high quality can be got from the model, which is much smaller the the source dataset.Though using the smaller one for visualization, it will take less time.Meanwhile; this model is supportive for varity kinds of interacton operations. In this paper, these problems will be solved in the following steps: First of all, this paper proposes a data cube visualization model for the three-dimensional scatter plot.The designing idea is creating a data cube space which be divided into many equal meta-data cubes for the visualization datasets.And each meta-data cube is a node of cube tree which is used for searching for new space and some other operations to speed up the data view processing. Then the distribution of local datasets can be computed by data cube visualization model with getting destiny estimation method. Then we can get a high-quality sample by using this model and improved the density biased sampling algorithm, which is main acceleration strategy on this data view prototype system to pursue the Effectiveness and efficiency.Finally, this paper has designed and implemented the three-dimensional scatter plot prototype system based on the data cube visualization model which is a speeding up system and has a variety of interactive operations. By analyzing experimental data and result, the prototype system has an improved functionality and efficiency comparing with other systems.