The Algorithm of Abstract Extraction Based on Semantics

When data mining involves document processing, extracting abstract information from the content of documents has become an essential procedure. The core idea of the algorithms of abstract extraction represented by Luhn is extracting the abstract merely from the sentences which contain frequent words in the essay. However, these algorithms fail to extract from the full text in a deeper semantic level. Therefore, the accuracy of the traditional abstract extraction algorithm needs to be enhanced. In order to improve the accuracy, we propose a method which can improve the performance of the algorithm of candidate key words extraction by using the substitute words and considering the semantic meanings of the candidate key words.