Text summarization using Natural language processing

It would be great to find a way to create a decent summary of a given text to use as a starting point for editing.

I think supplying this tool to an editor of a website, could drastically improve the experience of creating search and sharing-oriented summaries.

They would input the main text as normal, click a ‘generate summary’-button and get a summary of the desired lenght.

For instance 160 characters for description, 80 characters for title.

The editor could then edit the computer generated summary and create something in the right tone of voice, or adjust the robotic vibe before publishing.

So I think of this more like an assistive technology rather than a fully-automatic SEO beast.

Resources

https://github.com/icoxfog417/awesome-text-summarization#basic-approach

https://www.kdnuggets.com/2019/01/approaches-text-summarization-overview.html

https://www.nltk.org/

https://github.com/web64/norwegian-nlp-resources

http://vectors.nlpl.eu/repository/

https://spacy.io/models/nb

https://medium.com/analytics-vidhya/text-summarization-using-spacy-ca4867c6b744

https://www.facebook.com/groups/nlpnorway/

Approaches

Write a short summary of the different approaches including pros and cons here.