A Language Independent Method to ExtractEssence of a Text in the Form of phrases

In this paper, we present a method to extract essence of a text by statistical calculations. Essence of a text presents the most important concepts of that text and is used to retrieve documents in search engines more efficiently and to summarize the text more precisely. The essence is shown in the form of some phrases in this paper. So detection of main phrases of a text plays an important role in this method. We propose a language-independent method based on combination of statistical information extracted from document. In this method, there is no need to any natural language processing, neither ontologies nor document corpuses. We illustrate a real time method to understand main points of a text without any training data. Several parameters have been calculated for each phrase based on its frequency, frequency of related phrases and location of phrase in the text. Based on these parameters, some phrases are removed because of being unimportant. Remaining phrases presents the main points of the text. Evaluations show that the proposed approach provides a high precision and accuracy when extracts the essence of a single text in the form of phrases.

کلیدواژه ها:

essence of text ، text main points ، text processing ، language independent ، essence phrases