DeepEmotex: Classifying Emotion in Text Messages using Deep Transfer Learning (original) (raw)
Related papers
Textual emotion detection utilizing a transfer learning approach
The Journal of Supercomputing
Many attempts have been made to overcome the challenges of automating textual emotion detection using different traditional deep learning models such as LSTM, GRU, and BiLSTM. But the problem with these models is that they need large datasets, massive computing resources, and a lot of time to train. Also, they are prone to forgetting and cannot perform well when applied to small datasets. In this paper, we aim to demonstrate the capability of transfer learning techniques to capture the better contextual meaning of the text and as a result better detection of the emotion represented in the text, even without a large amount of data and training time. To do this, we conduct an experiment utilizing a pre-trained model called Emotional-BERT, which is based on bidirectional encoder representations from transformers (BERT), and we compare its performance to RNN-based models on two benchmark datasets, with a focus on the amount of training data and how it affects the models' performance.
Proceedings of the 13th International Workshop on Semantic Evaluation
Emotion detection from user-generated contents is growing in importance in the area of natural language processing. The approach we proposed for the EmoContext task is based on the combination of a CNN and an LSTM using a concatenation of word embeddings. A stack of convolutional neural networks (CNN) is used for capturing the hierarchical hidden relations among embedding features. Meanwhile, a long short-term memory network (LSTM) is used for capturing information shared among words of the sentence. Each conversation has been formalized as a list of word embeddings, in particular during experimental runs pre-trained Glove and Google word embeddings have been evaluated. Surface lexical features have been also considered, but they have been demonstrated to be not usefully for the classification in this specific task. The final system configuration achieved a micro F1 score of 0.7089. The python code of the system is fully available at https://github. com/marcopoli/EmoContext2019.
EmoDet at SemEval-2019 Task 3: Emotion Detection in Text using Deep Learning
Proceedings of the 13th International Workshop on Semantic Evaluation
Task 3, EmoContext, in the International Workshop SemEval 2019 provides training and testing datasets for the participant teams to detect emotion classes (Happy, Sad, Angry, or Others). This paper proposes a participating system (EmoDet) to detect emotions using deep learning architecture. The main input to the system is a combination of Word2Vec word embeddings and a set of semantic features (e.g. from AffectiveTweets Wekapackage). The proposed system (EmoDet) ensembles a fully connected neural network architecture and LSTM neural network to obtain performance results that show substantial improvements (F1-Score 0.67) over the baseline model provided by Task 3 organizers (F1score 0.58).
deepCybErNet at EmoInt-2017: Deep Emotion Intensities in Tweets
Proceedings of the 8th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, 2017
This working note presents the methodology used in deepCybErNet submission to the shared task on Emotion Intensities in Tweets (EmoInt) WASSA-2017. The goal of the task is to predict a real valued score in the range [0-1] for a particular tweet with an emotion type. To do this, we used Bag-of-Words and embedding based on recurrent network architecture. We have developed two systems and experiments are conducted on the Emotion Intensity shared Task 1 data base at WASSA-2017. A system which uses word embedding based on recurrent network architecture has achieved highest 5 fold crossvalidation accuracy. This has used embedding with recurrent network to extract optimal features at tweet level and logistic regression for prediction. These methods are highly language independent and experimental results shows that the proposed methods is apt for predicting a real valued score in than range [0-1] for a given tweet with its emotion type.
An Optimized Deep Learning Model for Emotion Classification in Tweets
Computers, Materials & Continua
The task of automatically analyzing sentiments from a tweet has more use now than ever due to the spectrum of emotions expressed from national leaders to the average man. Analyzing this data can be critical for any organization. Sentiments are often expressed with different intensity and topics which can provide great insight into how something affects society. Sentiment analysis in Twitter mitigates the various issues of analyzing the tweets in terms of views expressed and several approaches have already been proposed for sentiment analysis in twitter. Resources used for analyzing tweet emotions are also briefly presented in literature survey section. In this paper, hybrid combination of different model's LSTM-CNN have been proposed where LSTM is Long Short Term Memory and CNN represents Convolutional Neural Network. Furthermore, the main contribution of our work is to compare various deep learning and machine learning models and categorization based on the techniques used. The main drawback of LSTM is that it's a timeconsuming process whereas CNN do not express content information in an accurate way, thus our proposed hybrid technique improves the precision rate and helps in achieving better results. Initial step of our mentioned technique is to preprocess the data in order to remove stop words and unnecessary data to improve the efficiency in terms of time and accuracy also it shows optimal results when it is compared with predefined approaches.
How emotional are you? Neural Architectures for Emotion Intensity Prediction in Microblogs
2018
Social media based micro-blogging sites like Twitter have become a common source of real-time information (impacting organizations and their strategies, and are used for expressing emotions and opinions. Automated analysis of such content therefore rises in importance. To this end, we explore the viability of using deep neural networks on the specific task of emotion intensity prediction in tweets. We propose a neural architecture combining convolutional and fully connected layers in a non-sequential manner - done for the first time in context of natural language based tasks. Combined with lexicon-based features along with transfer learning, our model achieves state-of-the-art performance, outperforming the previous system by 0.044 or 4.4% Pearson correlation on the WASSA’17 EmoInt shared task dataset. We investigate the performance of deep multi-task learning models trained for all emotions at once in a unified architecture and get encouraging results. Experiments performed on eval...
Proceedings of the Fifth Workshop on Natural Language Processing and Computational Social Science (NLP+CSS)
In recent years, there has been increased interest in building predictive models that harness natural language processing and machine learning techniques to detect emotions from various text sources, including social media posts, micro-blogs or news articles. Yet, deployment of such models in real-world sentiment and emotion applications faces challenges, in particular poor out-of-domain generalizability. This is likely due to domainspecific differences (e.g., topics, communicative goals, and annotation schemes) that make transfer between different models of emotion recognition difficult. In this work we propose approaches for text-based emotion detection that leverage transformer models (BERT and RoBERTa) in combination with Bidirectional Long Short-Term Memory (BiLSTM) networks trained on a comprehensive set of psycholinguistic features. First, we evaluate the performance of our models within-domain on two benchmark datasets: GoEmotion (Demszky et al., 2020) and ISEAR (Scherer and Wallbott, 1994). Second, we conduct transfer learning experiments on six datasets from the Unified Emotion Dataset (Bostan and Klinger, 2018) to evaluate their out-of-domain robustness. We find that the proposed hybrid models improve the ability to generalize to out-of-distribution data compared to a standard transformer-based approach. Moreover, we observe that these models perform competitively on in-domain data.
Emotion Detection using CNN-LSTM based Deep Learning Model on Tweet Dataset
CERN European Organization for Nuclear Research - Zenodo, 2022
Emotion recognition from text is an important application of natural language processing. It has vast potential in many fields like marketing, artificial intelligence, political science, psychology etc. In recent times, more attention has been brought to this field because of availability and access to large amounts of opinionated data. Over the years many techniques have been proposed to tackle this problem. This paper focuses on the problem of emotion recognition from a dataset containing labelled tweets using a CNN-LSTM classifier model. The feature encoding for this model was done using the pre-trained Word2Vec word embedding and the model classified the tweets into five emotion classes: anger, sadness, joy, fear and love. The classifier was trained on 80% of the dataset and tested on the remaining 20%. The results of this proposed system was then compared with results from Multinomial Naïve Bayes (MNB), Support Vector Machine (SVM) and Convolution Neural Network (CNN) models. The proposed system was found to outperform all of them with an accuracy of 93.3%.
The paper describes the best performing system for EmoInt-a shared task to predict the intensity of emotions in tweets. Intensity is a real valued score, between 0 and 1. The emotions are classified as-anger, fear, joy and sadness. We apply three different deep neural network based models, which approach the problem from essentially different directions. Our final performance quantified by an average pearson correlation score of 74.7 and an average spearman correlation score of 73.5 is obtained using an ensemble of the three models. We outperform the base-line model of the shared task by 9.9% and 9.4% pearson and spearman correlation scores respectively.
Leveraging distant supervision and deep learning for twitter sentiment and emotion classification
Journal of intelligent information systems, 2024
Nowadays, various applications across industries, healthcare, and security have begun adopting automatic sentiment analysis and emotion detection in short texts, such as posts from social media. Twitter stands out as one of the most popular online social media platforms due to its easy, unique, and advanced accessibility using the API. On the other hand, supervised learning is the most widely used paradigm for tasks involving sentiment polarity and fine-grained emotion detection in short and informal texts, such as Twitter posts. However, supervised learning models are data-hungry and heavily reliant on abundant labeled data, which remains a challenge. This study aims to address this challenge by creating a largescale real-world dataset of 17.5 million tweets. A distant supervision approach relying on emojis available in tweets is applied to label tweets corresponding to Ekman's six basic emotions. Additionally, we conducted a series of experiments using various conventional machine learning models and deep learning, including transformer-based models, on our dataset to establish baseline results. The experimental results and an extensive ablation analysis on the dataset showed that BiLSTM with FastText and an attention mechanism outperforms other models in both classification tasks, achieving an F1-score of 70.92% for sentiment classification and 54.85% for emotion detection.