An Approach to Data Filtering Based on Fuzzy Correlation Space
Qian Xue · Chinese Journal of Computers · 2002
This paper presents a new approach to data filtering applied in large scale virtual environment. The structure of the paper is as follows: First, the drawbacks of traditional data filtering methods based on region dividing and multicast are analysed.There are many approaches in solving this problem (i.e. handling unclear information), such as AI and NN. In this paper, an approach based on fuzzy mathematics is used. It is proved that this approach is both simple and effective as well. With the establishment of fuzzy correlation space in VE based on the concept of entity correlation, the problem of data filtering is transformed into the problem of seeking correlative entity set in VE. Three data filtering algorithms are given in this paper, and the algorithm is used with self reliant threshold for the purpose of robustness. Finally, it is pointed out through simulation and theoretical analysis that this new approach can solve a series of problems that face traditional techniques, and improve both efficiency and robustness of data filtering in VE.