What is it about?
Lightweight semantic techniques are being researched to demonstrate the efficacy of summarizing multiple related news articles from the NIST Text Analysis Conference competition by simplifying sentences into Subject-Verb-Object representations, clustering those SVOs, and rating and selecting them to rank the top 5-6 sentences for a paragraph summary that beats the NIST baseline.
Featured Image
Photo by Joshua Hoehne on Unsplash
Why is it important?
Provides research into lightweight and portable machine-learning techniques for summarizing multiple related text documents into one short paragraph.
Perspectives
I hope this research inspires others to build more lightweight, portable text analysis techniques requiring little to no training. It would be great to see how this research can be built upon to transition from extractive to abstractive summarization.
Principal/Sr Staff Software Engineer Quinsulon Israel
Association for Computing Machinery
Read the Original
This page is a summary of: Semantic analysis for focused multi-document summarization (fMDS) of text, April 2015, ACM (Association for Computing Machinery),
DOI: 10.1145/2695664.2695672.
You can read the full text:
Contributors
The following have contributed to this page
