I spend a lot of time on google scholar making sure I haven't missed relevant literature. Usually this means scrolling through the "cited by" list one by one for at least 5-10 major papers on the topic.

It occurs to me that this could be done better and faster by a computer. Has anyone made software that does something like the following?

Input: A collection of papers

Output: Some papers similar to the input papers, based on citations or semantic analysis

I have in mind something like the "related papers" link on google scholar, but accepting more than one paper as input.

  • 1
    Mendeley tries this. – henning -- reinstate Monica Sep 4 '20 at 16:00
  • Thanks! This helps. It sounds like they have the methods I already use, plus Mendeley suggest (which I'll now try out). It's really good to learn about this. – oneguy Sep 4 '20 at 19:42

You may want to have a look at https://www.meta.org/ - you can create research interests and the search engine then suggests relevant articles for you (or tries to), based on machine learning.

Not the answer you're looking for? Browse other questions tagged or ask your own question.