A hybrid approach for learning concept hierarchy from Malay text using artificial immune network
Natural Computing: an international journal
| Hi-index | 0.00 |
Based on the basic theory of the representative artificial immune network model, a hierarchical clustering algorithm aiNHCA was presented. Firstly, aiNHCA generates the memory matrix and the similarity matrix of the antibodies in the aiNet method. So it can divide the data set into several sub-clusters. Then it combines the sub-clusters with higher similarities in the hierarchical clustering method and gets the final clustering result. The algorithm can be used in clustering the arbitrary shapes of data sets, and it has the advantages of fast discovering speed and high efficiency, which inherit the advantages of the immune algorithms. Using aiNHCA, the basic courses of researching on coal-gas outburst clustering of experimental samples, are analyzed based on artificial immune network model by using 5 indexes. At last, the coal-gas outburst clustering based on artificial immune network model by using synthetic index system including level, initial velocity of gas, crustal stress, comprehensive index, coal-seam thickness and tectonic complexity modulus is researched. The results of the experiment show that the prediction achieves an actuality level.