Closed maytepenella closed 6 years ago
I got worse results than with NB, but let's try.
Yes, me too. The teacher said that with SVM we need to introduce a Vector with less characteristics than RandomForest, if not, the result will be worst.
I reopen the issue because it could work if we use embedding for characteristic extraction and SVM as classifier. --> it is explained in reference about word2vec.
Add a optimization function in order to obtain best parameters for the model
I close this issue as it has already been implemented. And I open a new one about model optimization
Implement a SVM