SUMMARISATION OF LONG TEXT EXTRACTED FROM ARTICLE IMAGES BY INTEGRATING EXTRACTIVE AND ABSTRACTIVE TEXT SUMMARISATION METHODS

Jayaraj Balagopal

Abstract


People spend too much time reading huge articles, gist of which might be quite small. They also have tendencies to skip articles with essential content leading to not properly understanding the idea that the writer presents. Some articles also have a complex word which makes the articles difficult to perceive. This paper aims at extracting text from images of articles and summarising the latter using concepts in machine learning. The paper also identifies complex words from the document and substitutes simple words for the same. Sentence reconstruction is also performed to shorten long sentences to increase the scale of summarisation. The user can take images of the article he wishes to shorten and upload it to the server. The proposed application will extract the text from the image and provide the summarised version of the article to the user.

 

Keywords: Extractive summarisation, Abstractive summarisation, Parzen-window density function, LexRank, Encoder-Decoder Model, Attention Mechanism


Full Text:

PDF

Refbacks

  • There are currently no refbacks.


© 2017 International Journal of Global Research in Computer Science (JGRCS)
Copyright Agreement & Authorship Responsibility