I am working with tasks made by some occupations and am trying to find out the importance of these tasks within the occupation. My solution was to use tf-idf and then text rank and use word2vec and text rank. I would like to know if you guys have information about another ranking method besides text rank and if there is a comparison of the accuracy of these models in relation to text rank. Occupations do not have the same number of tasks The data is from O*NET( Occupational Information Network ). A database that contains hundreds of job definitions enter link description here
Asked
Active
Viewed 25 times
0
-
I think you might need to add more context to this question to get a meaningful answer. You mention occupations and occupational tasks, but not where you are getting copy to analyze, and if it is consistent across occupations (in quality, context and volume) These things matter. Try to add more information about the documents themselves. – sconfluentus Oct 05 '22 at 00:51