Review
Mathematics
Dehong Zeng, Xiaosong Chen, Zhengxin Song, Yun Xue, Qianhua Cai
Summary: An increasing number of people use different modalities to convey their opinions. In this work, we propose a novel Multimodal Interactive and Fusion Graph Convolutional Network for document-level multimodal sentiment analysis. The model incorporates text, images, and image captions to enhance semantics delivery and filter visual noise. Experimental results on a multimodal dataset show satisfying performance in sentiment analysis tasks.
Review
Computer Science, Artificial Intelligence
Ping Wang, Jiangnan Li, Jingrui Hou
Summary: This paper proposes a sentence-to-sentence attention network (S2SAN) using multihead self-attention for sentiment analysis, which outperforms other state-of-the-art models. By modeling attention at the sentence level, it enhances the accuracy of sentiment classifiers.
DECISION SUPPORT SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
You Zhang, Jin Wang, Xuejie Zhang
Summary: The proposed interactive attributes attention model considers user and product information in customer reviews, and utilizes a bilinear interaction and multiloss objective function to align attribute features with text representations for improved sentiment polarity classification performance.
KNOWLEDGE-BASED SYSTEMS
(2021)
Review
Computer Science, Information Systems
Yuehua Zhao, Linyi Zhang, Chenxi Zeng, Wenrui Lu, Yidan Chen, Tao Fan
Summary: This study utilizes a double-layer domain ontology for aspect-level sentiment analysis of online medical reviews. A double-layer aspect recognition model is built, and an object-aspect-sentiment knowledge graph is constructed, providing reference and guidance to sentiment analysis research in the online medical review domain.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Computer Science, Information Systems
Tian Shi, Xuchao Zhang, Ping Wang, Chandan K. Reddy
Summary: This study introduces an explanation method that captures causal relationships between keywords and model predictions by learning the importance of keywords for predicted labels across a training corpus based on attention weights. It can automatically learn higher-level concepts and their importance to model prediction tasks.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2022)
Review
Computer Science, Artificial Intelligence
Wei Zhenlin, Wang Chuantao, Yang Xuexin, Zhao Wei
Summary: This article presents a method to address the issue of imbalanced sample distribution in sentiment classification of online reviews. By training and using the SimBERT model in the experiment, fake samples are generated and mixed with the original samples to obtain a balanced dataset, thereby improving the classification performance of the model.
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
(2023)
Article
Business
Chia-Hsuan Chang, San-Yih Hwang, Ming-Lun Wu
Summary: High-quality sentiment lexicons are crucial for lexicon-based sentiment analysis, but most lexicons are only available in certain dominant languages, limiting their applicability in specific domains or languages. This paper proposes a multistep approach for bilingual sentiment lexicon induction to disambiguate words with opposite sentiment polarities, which outperforms existing lexicons and competing approaches in terms of accuracy and coverage, using experiments on real-world online review data sets.
ELECTRONIC COMMERCE RESEARCH AND APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Jaehun Park
Summary: Determining how to measure customer satisfaction with service quality and its determinants is a concern for practitioners and researchers. This study proposes a new methodology that combines data envelopment analysis (DEA) with text mining to analyze online textual reviews. The methodology identifies satisfaction metrics from online reviews using a TF-IDF algorithm, quantifies them by sentiment analysis, and evaluates service providers' LCS using a DEA model. An empirical case study applying this approach to the world's top 20 airlines in 2020 demonstrates its effectiveness.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Faliang Huang, Changan Yuan, Yingzhou Bi, Jianbo Lu, Liqiong Lu, Xing Wang
Summary: This paper proposes a probabilistic model for joint sentiment topic detection in online reviews, which outperforms state-of-the-art unsupervised approaches in sentiment detection quality and topic extraction ability, as demonstrated through experiments on seven sentiment analysis datasets.
APPLIED INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Guoshuai Zhao, Yiling Luo, Qiang Chen, Xueming Qian
Summary: This article presents a multitask learning model that combines ATE and APC tasks to extract aspect terms and classify aspect polarity simultaneously. It also utilizes multihead attention (MHA) to associate dependency sequences with aspect extraction, focusing on closely related words to aspects.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Yang Zhao, Tommy W. S. Chow
Summary: With the prosperity of online social media, there has been a significant increase in user-generated reviews. This paper introduces a sentiment subset selection framework to filter irrelevant sentiment information and select subsets based on topic modeling and submodular maximization with a cardinality constraint. Empirical analysis shows that the proposed framework can compress sentiment corpus while maintaining classifier performance on different metrics.
NEURAL COMPUTING & APPLICATIONS
(2021)
Review
Computer Science, Artificial Intelligence
Alexander Ligthart, Cagatay Catal, Bedir Tekinerdogan
Summary: This paper presents the results of a tertiary study on sentiment analysis, providing a comprehensive overview of key topics, different approaches, challenges, and unresolved issues in the field. In addition, recent 112 deep learning-based sentiment analysis papers were identified and analyzed based on the deep learning algorithms used.
ARTIFICIAL INTELLIGENCE REVIEW
(2021)
Article
Computer Science, Artificial Intelligence
Felipe Bravo-Marquez, Cristian Tamblay
Summary: This paper discusses three popular application domains of sentiment and emotion analysis: automatic rating of movie reviews, extracting opinions and emotions on Twitter, and inferring sentiment and emotion associations of words. The study proposes a method for transferring affective knowledge between words, tweets, and movie reviews using Word2Vec static embeddings and BERT contextualized embeddings, and finds that affective knowledge transfer is successful among the three domains.
COGNITIVE COMPUTATION
(2022)
Review
Computer Science, Information Systems
Yu Ji, Wen Wu, Yi Hu, Xi Chen, Jiayi Chen, Wenxin Hu, Liang He
Summary: This research proposes a Personality-Assisted Mood for Sentiment Classification (PAMSC) model that takes into account user mood for sentiment expression classification. The experimental results demonstrate that the PAMSC model achieves higher accuracy and interpretability compared to related models.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Jie Chen, Jingying Yu, Shu Zhao, Yanping Zhang
Summary: This study introduces a novel sentiment classification algorithm HUSN, which combines user review habits with hierarchical neural networks to enhance document-level sentiment classification performance. Experimental results demonstrate that the similarities between different reviews from the same user can improve sentiment classification effectiveness.
NEURAL PROCESSING LETTERS
(2021)