Automatic Text Documents Summarization through Sentences Clustering

To provide minimum redundancy in summary and maximum possible covering of the document content one has proposed the method of automatic summarization of the text documents based on sentence clustering. Sentence clustering is applied to determining topics and informative sentences. The clustering problem is formulated as a binary quadratic interger programming problem. The number of clusters (topics) is determined by deliberately developed algorithm. The synthesis algorithm of neural net for solving clustering problem has been described.