A real-time clustering algorithm based on Web usage mining;

Applying fuzzy cluster algorithm to Web information retrieval;

An improved K-means cluster algorithm

A learning algorithm based on cluster algorithm is presented and the ratio of recognition gets improved a lot.

An improved fuzzy clustering algorithm;

Scan line seed fill based clustering algorithm for two-value image;

Strategy-making method of overhead line control and undergroundcable retrofitting based on clustering algorithm;

To enhance the accuracy of hierarchical clustering algorithm,based on a study of space-constrained application,and using the Single Link algorithm as an example,the impact of space-constrainment is certificated to the result.

Support vector machines(SVMs) expert has a two-stage neural network architecture of SVMs,In the first stage,self-organizing feature map(SOM) is used as a clustering algorithm to partition the whole load input data into several disjointed regions,and the training data with similar features in the input space are belonged to the same region.

The emphasis of the thesis is to propose a model of image-meta search engines,and a vectorization method is adopted to apply HACM(hierarchical agglomerative clustering methods) clustering techniques on images search that are then optimized by a specially designed genetic algorithm.