In order to enable clustering to be done under a lower dimension, a new feature selection method for clustering is proposed. This method has three steps which are all carried out in a wrapper framework. First, all the original features are ranked according to their importance. An evaluation function E（f） used to evaluate the importance of a feature is introduced. Secondly, the set of important features is selected sequentially. Finally, the possible redundant features are removed from the important feature subset. Because the features are selected sequentially, it is not necessary to search through the large feature subset space, thus the efficiency can be improved. Experimental results show that the set of important features for clustering can be found and those unimportant features or features that may hinder the clustering task will be discarded by this method.
In order to improve the efficiency of regression testing in web application,the control flow graph and the greedy algorithm are adopted.This paper considers a web page as a basic unit and introduces a test case selection method for web application regression testing based on the control flow graph.This method is safe enough to the test case selection.On the base of features of request sequence in web application,the minimization technique and the priority of test cases are taken into consideration in the process of execution of test cases in regression testing for web application.The improved greedy algorithm is also raised resulting in optimization of execution of test cases.The experiments indicate that the number of test cases which need to be retested is reduced,and the efficiency of execution of test cases is also improved.
In order to reduce knowledge reasoning space and improve knowledge processing efficiency, a framework of distributed attribute reduction in concept lattices is presented. By employing the idea similar to that of the rough set, the characterization of core attributes, dispensable attributes and unnecessary attributes are described from the point of view of local formal contexts and virtual global contexts. A determinant theorem of attribute reduction is derived. Based on these results, an approach for distributed attribute reduction is presented. It first performs reduction independently on each local context using the existing approaches, and then local reducts are merged to compute reducts of global contexts. An algorithm implementation is provided and its effectiveness is validated. The distributed reduction algorithm facilitates not only improving computation efficiency but also avoiding the problems caused by the existing approaches, such as data privacy and communication overhead.
