Web document summarisation using pointwise mutual information (PMI) from web resources
by Atul Kumar Srivastava; Dhiraj Pandey; Alok Aggarwal; Sunil Gupta
International Journal of System of Systems Engineering (IJSSE), Vol. 12, No. 4, 2022

Abstract: Nowadays, large amount of data is generated over the internet. It is impossible for the humans to summarise such large chunks of bytes. Therefore, to deal with such challenges, automatic text summarisation systems are deployed. Text Summarisation is the field of data mining that highlights the relevance of important text in a document. In this paper, we proposed a web-based text summarisation approach that generates good quality summary based on total pointwise mutual information (TPMI) scores of the sentences. A sample document from DUC dataset is used which is pre-processed for tokenisation, stop words removal and stemming operations. Based on the extracted words, the TPMI is estimated by calculating the pointwise mutual information (PMI) of the occurrences of words on web search engine. To provide evidence for the robustness of our proposed system, proposed approach is compared with the well-known text summarisation techniques based on sentence length and mean score. The results show that our method outperforms the other techniques by exhibiting best results for closest mean score and generating good quality summary on sentences of different length.

Online publication date: Tue, 03-Jan-2023

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of System of Systems Engineering (IJSSE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com