Run_Files_README.txt

Running these files:

We put all three datasets in the Dataset.zip file

For different Feature Extractors with SVM and two other ensembling models:

There are 10 files inside each of these folders. For example, when we use BoW as a feature extractor, there are 10 (3 from each dataset and one for at once) files inside that folder. We use the naming format as 'Bow-feature-boost/svm/vot-IMDB/pro/res' and here boost for boosting ensemble model, sim for SVM classifier, vot for voting ensemble model, and IMDB for IMDB movie review dataset, pro for product review dataset and res for restaurant review dataset. We use the same format for all others also.

If you want to run the files together then run "run_all.py", it will give al the results together. 


For Standalone Models:

There are 4 files inside of these standalone models. Here we also use the same naming format. If you want to run all together then run "run_all.py"


We use codecarbon to report the CO2 emission.


If you feel any difficulties to run these files, please contract with me through email: kamruzzaman1@usf.edu