Customer review sentiment classification is used by several industries to understand how their product is performing.
In this case study, we will perform sentiment analysis on movie review data o IMDB. then, we will train a neural network to classify the data into a positive and negative positive.
The steps involved in training the neural network are as follows:
- To train this neural network, we consider the existing data of the movie reviews and its sentiments (0 or 1).
- 2. We convert the text data into its word vectors and convert each of the reviews into one vector as seen in the previous session.
- 3. Each review will have one sentence word embedding vector that can be fed into the neural network to train it for its sentiment.
Let’s understand these steps in detail. In the next video, you will understand how are input data looks like:
You can download the code files from here. Refer to README.md present in the link for proper instructions on downloading the code files.
Let us tokenise and pad the text as we had done in the previous segment.
Now we have tokenised and padded the text. In the next segment, we will convert this tokenized and padded matrix into word embeddings of this text.