jaykumarr / osdm Goto Github PK

Hi,

I tried to reproduce the scores reported in the paper using the Tweets in the data folder (I guess it is the Ts dataset referred to in the paper?) and the same parameters as reported.

Unfortunately, I get very low scores for all the metrics and was wondering what might be the problem.
Did you use the same parameters for all datasets?
Is the code the same as you used for your final experiments reported in the paper?

Best,
Tilman

Hi Jay. Could you please provide the Reuters dataset that you used for your experiments? I downloaded the data from Kaggle website and tried to remove documents with more than one class (as instructed in the paper) but I cannot reproduce the results. Thanks.