Article
Computer Science, Artificial Intelligence
David Valle-Cruz, Vanessa Fernandez-Cortez, Asdrubal Lopez-Chau, Rodrigo Sandoval-Almazan
Summary: Investors are constantly monitoring the behavior of stock markets which is influenced by social media reactions and emotions, especially during pandemics. Through financial sentiment analysis of Twitter data and financial indices, it was found that the markets reacted 0 to 10 days after information was shared on Twitter during the COVID-19 pandemic and 0 to 15 days after during the H1N1 pandemic. A lexicon-based approach and correlation analysis using SenticNet were effective in detecting highly shifted correlations. The most influential Twitter accounts during the pandemic were found to have a high correlation between sentiments on Twitter and stock market behavior.
COGNITIVE COMPUTATION
(2022)
Article
Computer Science, Artificial Intelligence
Hanane Grissette, El Habib Nfaoui
Summary: The research introduces a novel encoding method for affective biomedical concepts using sentic computing and neural networks, which enables sentiment analysis based on patient narrative data. Tested on COVID-19 related self-reports, the approach demonstrated effectiveness in inferring emotional information related to medication subjects, showing a positive impact on public health.
COGNITIVE COMPUTATION
(2022)
Article
Computer Science, Information Systems
Mohamed A. Aboali, Islam Elmaddah, Hossam E. Abdelmunim
Summary: This paper presents a novel content-based image retrieval methodology that uses neural networks to perform feature composition in a uniform-known domain. The proposed architecture was tested using different ResNet models for extracting image features and showed promising results.
Article
Engineering, Electrical & Electronic
Hongguang Zhu, Chunjie Zhang, Yunchao Wei, Shujuan Huang, Yao Zhao
Summary: Effective and efficient image-text retrieval is a challenging problem due to the gap between vision and language modalities. This study proposes a trade-off approach that balances effectiveness and efficiency by introducing the ESA module and SEL method. Experimental results demonstrate the effectiveness of the proposed method and its advantages in performance and retrieval time compared to other methods.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Ke Wang, Shengjie Luo, Tao Chen, Jianbo Lu
Summary: This article proposes a salient visual place recognition method that combines image retrieval, semantic information, and saliency cues for accurate estimations. The method uses a novel formulation to combine local semantic features into global descriptors and predicts saliency descriptors to learn the representation of static objects. A late-fusion module is introduced to increase the stability of the descriptor.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Information Systems
Song Yang, Qiang Li, Wenhui Li, Xuan-Ya Li, Ran Jin, Bo Lv, Rui Wang, Anan Liu
Summary: Image-text retrieval is a vital task in computer vision that aims to connect cross-modality data. This article proposes a novel semantic completion and filtration (SCAF) method to address the challenge of incomplete text descriptions of images and improve the learning of relevant region-word pairs.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2023)
Review
Computer Science, Artificial Intelligence
Renan Vinicius Aranha, Cleber Gimenez Correa, Fatima L. S. Nunes
Summary: The study focuses on the impact of Affective Computing in promoting user engagement in computer applications, with a systematic literature review of available articles discussing emotion recognition techniques and challenges to be overcome in the field.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2021)
Article
Computer Science, Information Systems
Qianli Xu, Ana Garcia Del Molino, Jie Lin, Fen Fang, Vigneshwaran Subbaraju, Liyuan Li, Joo-Hwee Lim
Summary: Lifelog analytics is a new research area incorporating technologies such as machine learning and data analytics. The proposed SRM method addresses the issue of lifelog information access, with a computational framework to support various applications.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Xingxu Yao, Dongyu She, Haiwei Zhang, Jufeng Yang, Ming-Ming Cheng, Liang Wang
Summary: The paper introduces a method for processing affective images through adaptive deep metric learning, which enhances the recognition of emotional images by designing adaptive sentiment similarity loss and sentiment vector, while also proposing a unified multi-task deep framework.
IEEE TRANSACTIONS ON MULTIMEDIA
(2021)
Article
Psychology, Multidisciplinary
Kuo-Liang Huang, Sheng-Feng Duan, Xi Lyu
Summary: The study focuses on exploring the differences in seven major acoustic features during voice interaction in relation to gender and emotional states of the PAD model. Thirty-one females and thirty-one males were invited to record audio of various emotions, and the extracted parameters were analyzed using statistical software. The results showed variations in gender and emotional states among the acoustic features, providing a theoretical foundation for AI emotional voice interaction.
FRONTIERS IN PSYCHOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
Siqi Zhu, Chunmei Qing, Canqiang Chen, Xiangmin Xu
Summary: This paper proposes a new emotion-based image transfer algorithm named Emotional Generative Adversarial Network (EGAN) to deal with the limitations of previous models. The proposed method utilizes adversarial training to generalize more emotional features and achieve diverse outputs. Experimental results demonstrate its superiority over other state-of-the-art baselines.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Keyu Chen, Xu Yang, Changjie Fan, Wei Zhang, Yu Ding
Summary: The ability to perceive human facial emotions is important in intelligent human-computer interaction. Previous research on facial emotion recognition (FER) has mainly focused on basic emotions or abstract dimensions, neglecting the richness of emotion statements. This study proposes a solution to address the semantic richness issue in FER and introduces a facial emotion recognition framework.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2022)
Article
Computer Science, Theory & Methods
Jing Han, Zixing Zhang, Maja Pantic, Bjoern Schuller
Summary: The study aims to shift traditional isolated affective computing into a lifelong learning paradigm, namely continual affective computing, and explores the lifelong learning algorithm of elastic weight consolidation in audiovisual emotion recognition across different cultures. The empirical results demonstrate the remarkable performance of the introduced lifelong learning approach in most cases.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Robotics
Jiahao Yin, Xinyu Zhou, Huahui Xiao, Zhili Liu, Wei Li, Xue Li, Shengyin Fan
Summary: In this paper, a system called HAPOR is proposed to solve the fundamental problem of camera pose estimation in a known scene graph. It combines image retrieval and iterative pose optimization using a hierarchical relocalization approach. The system filters out foreground dynamic objects and repeating textures using an attention mechanism and employs a graph-based image retrieval system and co-visibility graph for pose initialization and subsequent optimization. The application of HAPOR to the ORB-SLAM2 system achieves state-of-the-art relocalization results.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2023)
Article
Engineering, Electrical & Electronic
Dongrui Wu, Bao-Liang Lu, Bin Hu, Zhigang Zeng
Summary: A brain-computer interface (BCI) allows direct communication between a user and a computer through the central nervous system. An affective BCI (aBCI) monitors and regulates the emotional state of the brain, which has various applications in human cognition, communication, decision-making, and health. This tutorial provides a comprehensive and up-to-date guide on aBCIs, covering basic concepts, components of a closed-loop aBCI system, representative applications, and challenges and opportunities in aBCI research and applications.
PROCEEDINGS OF THE IEEE
(2023)
Article
Computer Science, Artificial Intelligence
Mohammad AL-Smadi, Mahmoud M. Hammad, Sa'ad A. Al-Zboon, Saja AL-Tawalbeh, Erik Cambria
Summary: The increasing interactive content in the Internet has led to research on Aspect-Based Sentiment Analysis (ABSA) in order to understand sentiments and aspects of a product in user comments. A deep learning model based on Gated Recurrent Units (GRU) and features extracted using the Multilingual Universal Sentence Encoder (MUSE) was developed for aspect extraction and polarity classification. The proposed Pooled-GRU model achieved high F1 scores of 93.0% for aspect extraction and 90.86% for aspect polarity classification, outperforming the baseline model and related research methods.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Siddique Latif, Heriberto Cuayahuitl, Farrukh Pervez, Fahad Shamshad, Hafiz Shehbaz Ali, Erik Cambria
Summary: This article provides a comprehensive survey on the progress of deep reinforcement learning (DRL) in the audio domain. By examining research studies in areas such as speech and music, the article discusses the methods and applications of DRL, and highlights the challenges and open areas for future research in the audio domain.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Artificial Intelligence
Jinjie Ni, Tom Young, Vlad Pandelea, Fuzhao Xue, Erik Cambria
Summary: This article surveys deep learning based dialogue systems, comprehensively reviewing and analyzing the state-of-the-art research outcomes in this field. It discusses different models in terms of principles, characteristics, and applications, and explores task-oriented and open-domain dialogue systems as two streams of research. The article also reviews evaluation methods and datasets for dialogue systems, and identifies possible future research trends.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Physics, Multidisciplinary
Giuseppe Varone, Cosimo Ieracitano, Aybike Ozyuksel Ciftcioglu, Tassadaq Hussain, Mandar Gogate, Kia Dashtipour, Bassam Naji Al-Tamimi, Hani Almoamari, Iskender Akkurt, Amir Hussain
Summary: The development of reinforced polymer composite materials has significantly influenced the problem of shielding high-energy photons in industrial and healthcare facilities. Machine learning approaches can be used to assess the gamma-ray shielding behavior of composites and provide an alternative to theoretical calculations. In this study, a dataset was developed and machine learning models were used to replicate the gamma-ray shielding characteristics of concrete.
Article
Computer Science, Artificial Intelligence
Mauajama Firdaus, Asif Ekbal, Erik Cambria
Summary: This research proposes a multilingual multitask approach that improves intent accuracy and slot filling for three different languages. The experimental results show an improvement in both tasks for all languages.
INFORMATION FUSION
(2023)
Article
Computer Science, Artificial Intelligence
Jintao Wen, Dazhi Jiang, Geng Tu, Cheng Liu, Erik Cambria
Summary: Multimodal data is crucial for enhanced emotion recognition in conversation, but effectively fusing different modal features to understand contextual information is challenging. This work proposes a Dynamic Interactive Multiview Memory Network (DIMMN) model, which integrates interaction information and mines crossmodal dynamic dependencies for emotion recognition. Experimental results show that DIMMN achieves better performance compared to state-of-the-art methods.
INFORMATION FUSION
(2023)
Article
Computer Science, Artificial Intelligence
Yu Ma, Rui Mao, Qika Lin, Peng Wu, Erik Cambria
Summary: This study proposes a novel Multi-source Aggregated Classification (MAC) method for predicting stock price movements. MAC incorporates numerical features, market-driven news sentiments, and sentiments of related stocks to better represent real market sentiments. The method also utilizes a graph convolutional network to capture the effects of news from related companies on the target stock. Extensive experiments demonstrate that MAC outperforms state-of-the-art models in predicting stock price movements, Sharpe Ratio, and backtesting trading incomes.
INFORMATION FUSION
(2023)
Article
Computer Science, Information Systems
Qian Liu, Rui Mao, Xiubo Geng, Erik Cambria
Summary: This paper conducts a systematic empirical study on semantic matching in machine reading comprehension (MRC). It formulates a two-stage framework and compares different setups of semantic matching modules on four MRC datasets. The study finds that semantic matching improves the effectiveness and efficiency of MRC, especially for answering questions with noisy and adversarial context. Matching coarse-grained context to questions and using semantic matching modules is more effective than fine-grained context matching, such as sentences and spans. However, semantic matching decreases the performance on why questions, suggesting that it is more helpful for questions that can be answered by retrieving information from a single sentence.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Computer Science, Interdisciplinary Applications
Phuong Le-Hong, Erik Cambria
Summary: This paper introduces a semantics-aware approach to natural language inference, in which explicit lexical and concept-level semantics from knowledge bases are incorporated to improve inference accuracy of neural network models. The authors conduct an extensive evaluation of four models with different sentence encoders, and experimental results show that semantics-aware neural models achieve higher accuracy than those without semantics information. On average of the three strong models, the proposed approach improves natural language inference in different languages.
LANGUAGE RESOURCES AND EVALUATION
(2023)
Article
Chemistry, Analytical
Heqing Huang, Bing Zhao, Fei Gao, Penghui Chen, Jun Wang, Amir Hussain
Summary: This paper proposes a novel unsupervised learning framework for video anomaly detection in smart city surveillance applications, using a training model based on the Cloze Test strategy. By encoding motion and appearance information at an object level, the proposed method improves the accuracy of anomaly perception. Comparative experiments on benchmark datasets demonstrate high AUROC scores through the proposed method.
Article
Education & Educational Research
Iti Chaturvedi, Erik Cambria, Roy E. Welsch
Summary: Video conferencing enables synchronous communication in classrooms and stimulates learners with multi-sensory content. Constructive pedagogy is used to teach complex AI equations, allowing students to experiment with different problem-solving methods. Multiple-choice questions provide reliable and quick assessments of student skill levels. The Australian Computer Society accreditation ensures flexible teaching templates for each subject. Geographic constraints necessitate subordinate campuses to be affiliated with a main campus. Continuity in learning and assessments between different subjects is maintained through the concept of strands. Feedback from students using AI-based simulations showed challenges in understanding lectures and assignments, hence a Kahoot quiz was introduced to measure learning. Charts were used to aid students with vision or attention-related disorders in visually observing variables and analyzing in-depth. Real-world industry examples were incorporated into lectures to enhance employability.
EDUCATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Geng Tu, Bin Liang, Dazhi Jiang, Ruifeng Xu
Summary: This article proposes a knowledge selection framework called SKSEC that incorporates sentiment emotion and context to improve emotion recognition in conversations. By eliminating and refining external knowledge, the performance of the model can be effectively enhanced.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Rui Mao, Qian Liu, Kai He, Wei Li, Erik Cambria
Summary: With the breakthrough of large-scale pre-trained language model (PLM) technology, prompt-based classification tasks, such as sentiment analysis and emotion detection, have gained increasing attention. This study conducts a systematic empirical study on prompt-based sentiment analysis and emotion detection to investigate the biases of PLMs in affective computing.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Deeksha Varshney, Asif Ekbal, Erik Cambria
Summary: This paper focuses on the neural-based interactive dialogue system that aims to engage and retain humans in long-lasting conversations. It proposes a new neural generative model that combines step-wise co-attention, self-attention-based transformer network, and an emotion classifier to control emotion and knowledge transfer during response generation. The results from quantitative, qualitative, and human evaluation show that the proposed models can generate natural and coherent sentences, capturing essential facts with significant improvement over emotional content.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Kai He, Rui Mao, Yucheng Huang, Tieliang Gong, Chen Li, Erik Cambria
Summary: In this paper, a prompt-based contrastive learning method is proposed for few-shot NER tasks. The method leverages external knowledge to initialize semantic anchors and optimizes prompts and sentence embeddings with a proposed semantic-enhanced contrastive loss. The method outperforms traditional contrastive learning methods in few-shot scenarios and effectively addresses the issues in conventional methods.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)