Article
Computer Science, Artificial Intelligence
Denilson Alves Pereira
Summary: Research on sentiment analysis specifically targeting the Portuguese language still needs further advancement to leverage the language's specificities. This paper surveys the current state of the art works in this area, categorizing and describing approaches as well as supporting language resources.
ARTIFICIAL INTELLIGENCE REVIEW
(2021)
Article
Computer Science, Theory & Methods
Yixuan Ma, Zhenji Zhang, Deming Li, Mincong Tang
Summary: This study introduces a data-driven approach that combines natural language processing techniques with the conditional logit model to address the inflation problem in reputation systems. By pre-training a multiplicative long short-term memory neural network, the proposed model effectively estimates deflated reputation information and provides better market outcomes.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Computer Science, Artificial Intelligence
Santwana Sagnika, Bhabani Shankar Prasad Mishra, Saroj K. Meher
Summary: Opinion-mining involves analyzing opinions on various topics in text form, enhancing accuracy by using subjectivity detection to differentiate between subjective and objective text. The model utilizes a combination of CNN and LSTM deep learning models and an attention network for improved effectiveness.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Necip Gozuacik, C. Okan Sakar, Sercan Ozcan
Summary: This study offers a novel approach by utilizing opinion retrieval theme along with sentiment analysis from social media platforms to support decision-making in product analysis and development. Google Glass is chosen as a case study, and a multi-task deep neural network architecture is designed for training sentiment prediction and opinion detection tasks.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Sharad Verma, Ashish Kumar, Aditi Sharan
Summary: Targeted Sentiment Analysis aims to identify the sentiment of a specific target aspect in a given text, going beyond general sentiment classification tasks. Previous studies mainly use recurrent neural networks (RNN) or variants to predict target-specific sentiment polarity, but RNN's sequential processing nature limits parallelization and fails to fully leverage modern multicore architectures' potential. Additionally, these models often overlook the inherent linguistic perspective embedded in the text. This paper proposes a novel approach called MuCon (Multi-channel Convolution), which uses a simple yet effective convolutional neural network (CNN) model. MuCon accurately determines aspect-specific sentiment polarity by incorporating multiple channels dedicated to linguistic and statistical features. By incorporating linguistic knowledge into a statistical model, MuCon performs better and achieves comparable results to sophisticated state-of-the-art methods.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Sarah Omar Alhumoud, Asma Ali Al Wazrah
Summary: The amount of Arabic content created on websites and social media has significantly increased in the past decade, leading to a rise in studies using recurrent neural networks (RNNs) for Arabic sentiment analysis. These studies vary in the areas they address, the functionality and weaknesses of the models, and the number and scale of available datasets.
ARTIFICIAL INTELLIGENCE REVIEW
(2022)
Review
Computer Science, Cybernetics
Laxmi Chaudhary, Nancy Girdhar, Deepak Sharma, Javier Andreu-Perez, Antoine Doucet, Matthias Renz
Summary: This study introduces the importance of sentiment analysis on the microblogging site Twitter. It provides an overview of standard preprocessing techniques and word embeddings for data preparation, and offers a comprehensive summary of deep learning-based approaches. Additionally, the study compiles popular benchmark datasets and discusses domain-specific practical applications of sentiment analysis tasks. Finally, it concludes with research challenges and outlines future prospects for further investigation.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Tinghuai Ma, Huan Rong, Yongsheng Hao, Jie Cao, Yuan Tian, Mznah Al-Rodhaan
Summary: This research focuses on sentiment polarity detection from online user-generated text. The existing lexicon-based methods suffer from polarity fuzziness, where the same word can have opposite polarities in different seed lexicons. To address this issue, the study proposes a two-aspect lexicon expansion approach to enhance Chinese sentiment polarity detection. By detecting and revising sentiment polarity for new and existing words in seed lexicons and incorporating fine-grained sentiment processing through symmetrical mapping, sentiment feature pruning, and text representation, the proposed framework achieves the best overall performance compared to other methods.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Estela Saquete, Jose Zubcoff, Yoan Gutierrez, Patricio Martinez-Barco, Javi Fernandez
Summary: The focus of this research is to discover the main features of virality patterns in Twitter. Five trending topics related to the COVID-19 pandemic in Spanish language were selected. Opinion mining techniques were used to structure the information based on message polarity and emotions. Data mining techniques, specifically association rules mining, were then applied to identify the highest viral message patterns and the relevant characteristics of patterns with low impact. The analysis revealed that messages with high-negative polarity and intense emotions, particularly fear, sadness, anger, and surprise intensified by the COVID-19 pandemic, are more likely to go viral on social media. On the other hand, messages with little news coverage, few authors, and the absence of surprise are relevant features for messages with very low dissemination on social media.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Hardware & Architecture
Ghazi Abdalla, Fatih Ozyurt
Summary: In modern society, the Internet has become a basic necessity for people, and sentiment analysis using deep learning techniques enables companies to improve product quality.
Article
Economics
Anamika Anamika, Sowmya Subramaniam
Summary: This paper examines the influence of investor sentiment based on news headlines on the Cryptocurrency Market Index and individual cryptocurrency returns. The findings show that news sentiment has a significant impact on cryptocurrency returns, particularly for young, small, and volatile cryptocurrencies.
Article
Multidisciplinary Sciences
Sultan Zeybek, Duc Truong Pham, Ebubekir Koc, Aydin Secer
Summary: A novel metaheuristic optimization approach for training deep RNNs for sentiment classification task is proposed, demonstrating significant advantages. The BA-3+ algorithm outperforms DE and PSO algorithms in terms of speed and accuracy, surpassing SGD, DE, and PSO algorithms.
Review
Mathematics
Muhammad Shehrayar Khan, Atif Rizwan, Muhammad Shahzad Faisal, Tahir Ahmad, Muhammad Saleem Khan, Ghada Atteia
Summary: With the increase in users of social media websites such as IMDb and the availability of publicly accessible data, opinion mining has become more accessible. This study explores the categorization of movie reviews, which can be challenging due to the complexity of human language. The use of the Word2Vec model and various features, such as psychological, readability, and linguistic features, were investigated. The results showed that the SVM algorithm with self-trained Word2Vec achieved an F-Measure of 86%, while using a combination of psychological, linguistic, readability features, and Word2Vec features resulted in an F-Measure of 87.93%.
Article
Computer Science, Cybernetics
Prasoon Gupta, Sanjay Kumar, R. R. Suman, Vinay Kumar
Summary: This study analyzed the sentiments of Indian citizens towards the nationwide lockdown enforced by the Indian government during the COVID-19 outbreak using NLP and machine learning classifiers. The majority of Indian citizens were found to support the government's decision to implement the lockdown.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2021)
Article
Computer Science, Information Systems
Lili Shang, Meiyun Zuo
Summary: In this study, a fused sequential and hierarchical representation (FSHR) model is proposed for extracting aspect terms from opinionated sentences. The model combines sequential and hierarchical representations to capture both linear semantic information for predicting meaning-related aspect terms and syntactic relations for identifying structure-related aspect terms. Experimental results demonstrate that FSHR outperforms competitive baselines, and further analysis reveals the effectiveness of the model.
JOURNAL OF INFORMATION SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Olivier Habimana, Yuhua Li, Ruixuan Li, Xiwu Gu, Wenjin Yan
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2020)