Multi-Document Text Summarization of Marathi Regional Language
Author(s):
Aman Jain, Sujata Lungare, Tejashri Bhingare, Priyanka Tak, Mr .Pratik Kamble
Keywords:
Extractive text summarization , Text mining , Stop word removal , Bag of words
Abstract
In this new aeon , where enormous information is available on the internet, the best significance is to provide improved mechanism to extract information in a short period of time and efficiently . It is inconvenient for human beings to manually extract summary of large documents of text .There are ample text materials available on the internet , so it is not easy to find relevant documents from a number of documents which are currently available .In order to solve this problem Automated Text Summarization comes into picture .Text Summarization is a process of pinpointing the most important meaningful information in a document or a set of documents compressing them in to a shorter version conserving its overall meanings .In this paper the author has given a brief view of how a user will be using a web application to summarize Marathi articles or documents .Such summary is generated using probabilistic model for multi-source Marathi text summarization . Here, precision, recall , compression ratio as well as retention ratio are the parameters which are used for evaluation purpose .Implemented system successfully generates relevant summary of Marathi articles.