Sequential combination methods for data clustering analysis
Journal of Computer Science and Technology
An Evolutionary Immune Network for Data Clustering
SBRN '00 Proceedings of the VI Brazilian Symposium on Neural Networks (SBRN'00)
A divisive information theoretic feature clustering algorithm for text classification
The Journal of Machine Learning Research
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Solving mathematical problems using knowledge-based systems
Mathematics and Computers in Simulation - Special issue: Applications of computer algebra in science, engineering, simulation and special software
Mountain Clustering on Nonuniform Grids
AIPR '04 Proceedings of the 33rd Applied Imagery Pattern Recognition Workshop
Immune model-based fault diagnosis
Mathematics and Computers in Simulation
Fast accurate fuzzy clustering through data reduction
IEEE Transactions on Fuzzy Systems
Adaptive immune genetic algorithm for logic circuit design
Proceedings of the first ACM/SIGEVO Summit on Genetic and Evolutionary Computation
An immune co-evolutionary algorithm based approach for optimization control of gas turbine
Proceedings of the first ACM/SIGEVO Summit on Genetic and Evolutionary Computation
Hi-index | 0.00 |
Evaluation on the performance and quality of textile products is very important in textile industry, for example, clustering-based fabric evaluation. Classical clustering methods have some disadvantages, one of which is that the parameters of fabrics are straightly clustered without extracting their characteristics. As thus, it brings about a result that the computation efficiency depends on spatial distributing of data. In this paper, we introduce a novel immune-based evolutionary algorithm (IEA) to overcome this limitation. The IEA, inspired from the defending mechanism of biological immune system, has better capability of global searching and diversiform-memorizing. To explain that the IEA-based clustering method is superior to classical clustering ones, we first prove its better performance for clustering problem via two functions, and then apply it to fabric sample clustering. The sample data includes 43 fabrics with 12 KES parameters, which are self-knitted by Ecole Nationale Superieure des Arts et Industries Textiles, France. By iterative calculating, new center points can be obtained gradually according to the information learned from given sample data, and then the best clustering centers can be obtained. The significant innovation of the IEA for clustering fabric is that the sample characteristic is refined to be the center of the points in a group by iterative learning. Compared with classical clustering methods used for fabric evaluation, the IEA can learn and adapt to the structure of sample, and then find out characteristics with better clustering result. The simulation results demonstrate that the IEA can adapt to the non-balanced environment in a short time and recognize the learned object steadily and quickly.