Agriculture is the back-bone of economy in India. In order to enhance development in this field knowing only basics of agriculture is not enough. Researcher or framer should be aware of agriculture practices throughout the world. As English is global language most of the information is available in English, so when one browses agriculture information, it becomes difficult for some farmers to get the correct meaning of the information and quite difficult to go through each and every information. This result in language barrier and information overload that either leads to wastage of significant time browsing all information or else useful information is missed out. Hence text summarization in native language in agriculture field is very essential for user to get concise information about new technology. The proposed methodology comprises of machine translation, data pre-processing and automatic text summarization. The machine translation phase translates documents which are either in Hindi or Marathi language to the English document. After that data pre-processing takes place. Data pre-processing step involves noise removal, tokenization, stop word removal and stemming. On the pre-processed data automatic text summarization is performed using clustering approach. Then according to the user choice the summary will be translated to Hindi or Marathi language.

Keywords : Data mining, text summarization, extractive summarization, K-means clustering, Machine translation.