Sentiment Analysis in the Light of LSTM Recurrent Neural Networks

Abstract

Long short-term memory (LSTM) is a special type of recurrent neural network (RNN) architecture that was designed over simple RNNs for modeling temporal sequences and their long-range dependencies more accurately. In this article, the authors work with different types of LSTM architectures for sentiment analysis of movie reviews. It has been showed that LSTM RNNs are more effective than deep neural networks and conventional RNNs for sentiment analysis. Here, the authors explore different architectures associated with LSTM models to study their relative performance on sentiment analysis. A simple LSTM is first constructed and its performance is studied. On subsequent stages, the LSTM layer is stacked one upon another which shows an increase in accuracy. Later the LSTM layers were made bidirectional to convey data both forward and backward in the network. The authors hereby show that a layered deep LSTM with bidirectional connections has better performance in terms of accuracy compared to the simpler versions of LSTM used here.

Multimodal sentimental analysis for social media applications: A comprehensive review

“…The enhanced versions of RNN are gated recurrent units and long short term memory units (LSTM). To carry out the sentiment analysis of reviews the authors (Pal, Ghosh, & Nag, 2018) have applied the different LSTM architectures and they have shown that the RNN-LSTM is more effective than deep and CNNs. The performance of the Bi-LSTM model is superior to the other versions of LSTM.…”

Sentiment Prediction for User Comments on Home Appliances Products

“…Recently, document representation methods such as Doc2vec [13] where the model is trained to represent a sentence or even a paragraph, have been introduced to solve sentiment classification problems from document level, and it shows higher performance than bag-of-words approaches [14]. It is also noted that deep learning technology has been shown to be effective in a wide range of natural language processing tasks [15].…”

Automatic Identification Methods on a Corpus of Twenty Five Fine-Grained Arabic Dialects

“…Their main characteristic is that they remember information for long periods of time. This class of neural network has been efficient for many NLP tasks such as language modeling [24], sentiment analysis [16], word embedding learning [11], as well as in other area like automatic speech recognition [7] and image captioning [13].…”