I am working on a NLP dataset, and now I am mainly working on improving its accuracy, and I think I know why I am not getting high accuracy, that it is because of the two reasons:
- One is the Lemmatization techniques that I used on my dataset is not working properly because, nltk’s WordNetLemmatizer lemmatize only nouns, as it is the default one, so I couldn’t lemmatize other words, 2) Another reason is that there are too many slangs present in my text data, which I am not able to rectify.
Someone please suggest me some solution as to how to proceed forward, and get these two issues resolved.
1 post - 1 participant