如何使用nlp对我拥有的数据集将评论分为好与坏？

Question

the tsv file of the data Things i have tried on jupyter

I have this data of customer reviews for a clothing e commerce store, i am learning 'nlp' using python on 'jupyter-notebook' and i wanted to learn how to classify the 'Review Text' column's reviews as good or bad using nlp.

• List item

i have imported the file and separated with a delimiter cleaned the 'Review Text' column data tokenisation of the data such as converting to lower case , stopwords emitting, stemmering and split.

please do help me in this project. i have tried doing it by learning from a few blogs but it hasn't helped a lot.

Answer 1

By seeing your dataset I am assuming that you can take the Review Text column as independent variable and "Positive feedback " column as the dependent one which consists of 0's and 1's Step 1: Doing the stemming process for "Review text" column Step 2 :spliting the words and converting to lower and removing all regular expressions Step 3 : Use Count vectorizer Step 4 : Next train_test_split in x= "Review text" y = "Positive feedback" step 5 : Next use any classifier for classifying the words into 0 and 1

For further Guidance once see this link https://www.kaggle.com/apekshakom/sentiment-analysis-of-restaurant-reviews