Automatic Extraction of Idiom, Proverb and its Variations from Text using Statistical Approach

Natural languages are full of idiomatic uses, which while translating through present NLP system do not extract variations of idioms and proverbs. To overcome this problem, a new method to extract idioms/proverbs is proposed in this paper. The proposed methodology uses statistical method to automatically extract idioms and proverbs from the text along with their variations. The system is updated with a huge database of idioms and proverbs with all of their variations and then tested on a large text file of 'Panchatantra Tales'. The system gave an accuracy of more than 80%, which proves that our method is a successful approach in correctly interpreting and generating the translation of natural language.