Open abhinavtripathy opened 4 years ago
These are some of the papers I was scanning through, they are related to machine learning so it was kind of hard for me to understand their math and logic. I added links to two research papers and I also found two ranking programs you might want to look at. (Someone in the team with machine learning knowledge should check these out, there are a lot of ml things that was hard for me to understand)
http://precog.iiitd.edu.in/Publications_files/a2-gupta.pdf https://www.aclweb.org/anthology/C10-1034.pdf http://www.cs.cornell.edu/people/tj/svm_light/svm_rank.html#:~:text=SVMrank%20is%20an%20instance,of%20the%20folds%20and%20datasets. https://en.wikipedia.org/wiki/Learning_to_rank
Summary: One of the papers I found relevant was on analyzing the credibility of information in tweets corresponding to high impact news events of 2011 around the globe. They used linear logistic regression analysis to identify twitter features like the number of unique characters, swear words, pronounce that can help in assessing credibility. They evaluated an algorithm using RankSVM to rank the tweets according to their credibility score.
We need to research on what sort of analytics we can do in terms of tagging context, credibility of a source etc.