13.9. Text Sentiment Classification: Using Recurrent Neural Networks

Similar to search synonyms and analogies, text classification is also a downstream application of word embedding. In this section, we will apply pre-trained word vectors and bidirectional recurrent neural networks with multiple hidden layers [Maas et al., 2011]. We will use them to determine whether a text sequence of indefinite length contains positive or negative emotion. Import the required package or module before starting the experiment.

import d2l
from mxnet import gluon, init, nd
from mxnet.gluon import nn, rnn
from mxnet.contrib import text

batch_size = 64
train_iter, test_iter, vocab = d2l.load_data_imdb(batch_size)

13.9.1. Use a Recurrent Neural Network Model

In this model, each word first obtains a feature vector from the embedding layer. Then, we further encode the feature sequence using a bidirectional recurrent neural network to obtain sequence information. Finally, we transform the encoded sequence information to output through the fully connected layer. Specifically, we can concatenate hidden states of bidirectional long-short term memory in the initial time step and final time step and pass it to the output layer classification as encoded feature sequence information. In the BiRNN class implemented below, the Embedding instance is the embedding layer, the LSTM instance is the hidden layer for sequence encoding, and the Dense instance is the output layer for generated classification results.

class BiRNN(nn.Block):
    def __init__(self, vocab_size, embed_size, num_hiddens, num_layers, **kwargs):
        super(BiRNN, self).__init__(**kwargs)
        self.embedding = nn.Embedding(vocab_size, embed_size)
        # Set Bidirectional to True to get a bidirectional recurrent neural
        # network
        self.encoder = rnn.LSTM(num_hiddens, num_layers=num_layers,
                                bidirectional=True, input_size=embed_size)
        self.decoder = nn.Dense(2)

    def forward(self, inputs):
        # The shape of inputs is (batch size, number of words). Because LSTM
        # needs to use sequence as the first dimension, the input is
        # transformed and the word feature is then extracted. The output shape
        # is (number of words, batch size, word vector dimension).
        embeddings = self.embedding(inputs.T)
        # Since the input (embeddings) is the only argument passed into
        # rnn.LSTM, it only returns the hidden states of the last hidden layer
        # at different time step (outputs). The shape of outputs is
        # (number of words, batch size, 2 * number of hidden units).
        outputs = self.encoder(embeddings)
        # Concatenate the hidden states of the initial time step and final
        # time step to use as the input of the fully connected layer. Its
        # shape is (batch size, 4 * number of hidden units)
        encoding = nd.concat(outputs[0], outputs[-1])
        outs = self.decoder(encoding)
        return outs

Create a bidirectional recurrent neural network with two hidden layers.

embed_size, num_hiddens, num_layers, ctx = 100, 100, 2, d2l.try_all_gpus()
net = BiRNN(len(vocab), embed_size, num_hiddens, num_layers)
net.initialize(init.Xavier(), ctx=ctx) Load Pre-trained Word Vectors

Because the training data set for sentiment classification is not very large, in order to deal with overfitting, we will directly use word vectors pre-trained on a larger corpus as the feature vectors of all words. Here, we load a 100-dimensional GloVe word vector for each word in the dictionary vocab.

glove_embedding = text.embedding.create(
    'glove', pretrained_file_name='glove.6B.100d.txt')

Query the word vectors that in our vocabulary.

embeds = glove_embedding.get_vecs_by_tokens(vocab.idx_to_token)
(49339, 100)

Then, we will use these word vectors as feature vectors for each word in the reviews. Note that the dimensions of the pre-trained word vectors need to be consistent with the embedding layer output size embed_size in the created model. In addition, we no longer update these word vectors during training.

net.embedding.collect_params().setattr('grad_req', 'null') Train and Evaluate the Model

Now, we can start training.

lr, num_epochs = 0.01, 5
trainer = gluon.Trainer(net.collect_params(), 'adam', {'learning_rate': lr})
loss = gluon.loss.SoftmaxCrossEntropyLoss()
d2l.train_ch12(net, train_iter, test_iter, loss, trainer, num_epochs, ctx)
loss 0.425, train acc 0.812, test acc 0.821
907.5 exampes/sec on [gpu(0), gpu(1)]

Finally, define the prediction function.

# Save to the d2l package.
def predict_sentiment(net, vocab, sentence):
    sentence = nd.array(vocab[sentence.split()], ctx=d2l.try_gpu())
    label = nd.argmax(net(sentence.reshape((1, -1))), axis=1)
    return 'positive' if label.asscalar() == 1 else 'negative'

Then, use the trained model to classify the sentiments of two simple sentences.

predict_sentiment(net, vocab, 'this movie is so great')
predict_sentiment(net, vocab, 'this movie is so bad')

13.9.2. Summary

  • Text classification transforms a sequence of text of indefinite length into a category of text. This is a downstream application of word embedding.

  • We can apply pre-trained word vectors and recurrent neural networks to classify the emotions in a text.

13.9.3. Exercises

  • Increase the number of epochs. What accuracy rate can you achieve on the training and testing data sets? What about trying to re-tune other hyper-parameters?

  • Will using larger pre-trained word vectors, such as 300-dimensional GloVe word vectors, improve classification accuracy?

  • Can we improve the classification accuracy by using the spaCy word tokenization tool? You need to install spaCy: pip install spacy and install the English package: python -m spacy download en. In the code, first import spacy: import spacy. Then, load the spacy English package: spacy_en = spacy.load('en'). Finally, define the function def tokenizer(text): return [tok.text for tok in spacy_en.tokenizer(text)] and replace the original tokenizer function. It should be noted that GloVe’s word vector uses “-” to connect each word when storing noun phrases. For example, the phrase “new york” is represented as “new-york” in GloVe. After using spaCy tokenization, “new york” may be stored as “new york”.

13.9.4. Scan the QR Code to Discuss