Linguistically Regularized LSTM for Sentiment Classification

Qiao, Qiao; Huang, Minlie; Lei, Jinhao; Zhu, Xiaoyan

doi:10.18653/v1/p17-1154

Cited by 168 publications

(77 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Freebase) to learn knowledge representations for entity classification and knowledge graph completion. Qian et al (2016) propose linguistically regularized LSTMs for sentiment analysis with sentiment lexicons, negation words, and intensity words. In this work, we encode semantic features into convolutional layers by initializing them with important n-grams.…”

Section: Related Workmentioning

confidence: 99%

Initializing Convolutional Filters with Semantic Features for Text Classification

Shen¹,

Zhao²,

Liu³

et al. 2017

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) are widely used in NLP tasks. This paper presents a novel weight initialization method to improve the CNNs for text classification. Instead of randomly initializing the convolutional filters, we encode semantic features into them, which helps the model focus on learning useful features at the beginning of the training. Experiments demonstrate the effectiveness of the initialization technique on seven text classification tasks, including sentiment analysis and topic classification.

show abstract

Section: Related Workmentioning

confidence: 99%

Initializing Convolutional Filters with Semantic Features for Text Classification

Shen¹,

Zhao²,

Liu³

et al. 2017

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

show abstract

“…Meanwhile, Qian et al [37] utilized Long Short-Term Memory (LSTM) [38][39][40][41] for binary classification of sentiment and obtained 82.1% accuracy on the movie review data [27]. Kim [1] had a result of a maximum of 89.6% accuracy with seven different types of data through their CNN model with one convolutional layer.…”

Section: Deep Learning For Sentiment Classificationmentioning

confidence: 99%

Sentiment Classification Using Convolutional Neural Networks

2019

View full text Add to dashboard Cite

As the number of textual data is exponentially increasing, it becomes more important to develop models to analyze the text data automatically. The texts may contain various labels such as gender, age, country, sentiment, and so forth. Using such labels may bring benefits to some industrial fields, so many studies of text classification have appeared. Recently, the Convolutional Neural Network (CNN) has been adopted for the task of text classification and has shown quite successful results. In this paper, we propose convolutional neural networks for the task of sentiment classification. Through experiments with three well-known datasets, we show that employing consecutive convolutional layers is effective for relatively longer texts, and our networks are better than other state-of-the-art deep learning models.

show abstract

“…In [18] they propose simple, trained models with annotations where each sentence is annotated with two classes as negative, positive at the sentence level, and they also model the linguistic role of the lexical feelings with LSTM, the words of denial and the words of intensity. The results show that the models are capable of capturing the linguistic role of sentimental words, words of denial and words of intensity in sentimental expression.…”

Section: Sentiment Analysis Using Deep Learningmentioning

confidence: 99%

Emotion Recognition for Education using Sentiment Analysis

Barrón-Estrada¹,

Zataraín-Cabada²,

Bustillos³

2019

RCS

View full text Add to dashboard Cite

Most systems that recognize emotions that are used in learning systems are oriented towards the detection of basic emotions such as happy, sad or angry. However, an intelligent tutor system must be able to detect, in a student, secondary emotions that have to do with learning as boring or frustrated. In the case of sentiment analysis in text, most of the research focuses on recognizing polarities (positive and negative). In this paper we present a sentiment analyzer for the recognition of polarities and emotions centered on learning using textual phrases written in Spanish in the field of computer science that can be used in an ITS (or any learning environment) to detect sentiments and/or emotions of the students. We also present the results obtained from the sentiment analyzer, which are very promising with 88.26% of accuracy.

show abstract

Linguistically Regularized LSTM for Sentiment Classification

Cited by 168 publications

References 39 publications

Initializing Convolutional Filters with Semantic Features for Text Classification

Initializing Convolutional Filters with Semantic Features for Text Classification

Sentiment Classification Using Convolutional Neural Networks

Emotion Recognition for Education using Sentiment Analysis

Contact Info

Product

Resources

About