Article
Chemistry, Multidisciplinary
Qizhi Li, Xianyong Li, Yajun Du, Yongquan Fan, Xiaoliang Chen
Summary: This paper proposes a new sentiment-enhanced word embedding method to improve sentence-level sentiment classification. By leveraging the mapping relationship between word embeddings and sentiment orientations, the method achieves higher accuracy and F1 values and reduces convergence time in sentiment classification models.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Artificial Intelligence
Wazib Ansar, Saptarsi Goswami, Amlan Chakrabarti, Basabi Chakraborty
Summary: This paper proposes a novel Transformer encoder architecture with SLFN and PSCE to improve performance and efficiency. The proposed architecture achieves satisfactory results in sentiment analysis through improvements in reducing parameters and increasing accuracy.
APPLIED INTELLIGENCE
(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
Peng Wu, Xiaotong Li, Chen Ling, Shengchun Ding, Si Shen
Summary: The study proposes a sentiment classification method for large scale microblog text using the attention mechanism and bidirectional long short-term memory network. Through experimental comparison with baseline methods, the efficacy of the proposed method is demonstrated. The main novelty lies in the incorporation of the attention mechanism in a deep learning network for analyzing large scale social media data.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Rini Wijayanti, Andria Arisal
Summary: A novel Indonesian sentiment lexicon (SentIL) was created using an automatic pipeline, involving seed word creation, slang words and emoticons addition, and sentiment value tuning. Experimental results showed a significant increase in lexicon size and improved accuracy, outperforming other available Indonesian sentiment lexicons.
ACM TRANSACTIONS ON ASIAN AND LOW-RESOURCE LANGUAGE INFORMATION PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Qizhi Zhao, Zan Mo, Mengting Fan
Summary: This study proposes a model called POS-ATAEPE-BiLSTM for ABSA in Chinese. The model extracts aspect multiwords based on POS tagging and includes POS information in the embedding layer. The study demonstrates the effectiveness of the proposed model in Chinese ABSA tasks.
APPLIED INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Felipe Bravo-Marquez, Arun Khanchandani, Bernhard Pfahringer
Summary: This study introduces a method to automatically induce continuously updated sentiment lexicons by training incremental word sentiment classifiers from Twitter streams. Experimental results show that the approach allows for successful tracking of the sentiment of words over time.
COGNITIVE COMPUTATION
(2022)
Article
Computer Science, Artificial Intelligence
Yusi Qi, Xiaoqing Zheng, Xuanjing Huang
Summary: This article introduces a method for aspect term sentiment classification, which mainly solves the problem that the same word may express different sentiment polarities for different aspects. By creating aspect-sensitive lexicons and applying enhanced sentiment analysis models, it achieved state-of-the-art performances on multiple datasets.
KNOWLEDGE AND INFORMATION SYSTEMS
(2022)
Article
Acoustics
Yuanzhi Liu, Min He, Qingqing Yang, Gwanggil Jeon
Summary: Text sentiment style transfer aims to extract sentiment words and transfer them into another desired sentiment style while preserving the original sentence's content. A novel framework with attention mechanism and embedding perturbed encoder is proposed to enhance the performance of non-parallel text sentiment style transfer. The framework utilizes reverse attention to disentangle sentiment style information, embedding perturbation to add adjustable noise for better semantic representation, and attention mechanism to focus on high-weight words during decoding, resulting in improved sentiment style transfer accuracy, content preservation, and language fluency compared to previous works.
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
(2023)
Article
Computer Science, Artificial Intelligence
Zahra Rahimi, Mohammad Mehdi Homayounpour
Summary: Word embeddings as feature learning techniques for natural language processing have limitations in sentiment analysis, but the proposed unsupervised models that integrate word polarity and co-occurrence information show higher performance in document-level sentiment analysis tasks.
APPLIED INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Hanzheng Wang, Jiaqi Zhao, Yong Zhou, Rui Yao, Ying Chen, Silin Chen
Summary: The proposed attentive modality-consistent network (AMC-Net) for Visible-Infrared Person Re-Identification (VI-ReID) effectively mines spatial and channel information while narrowing down the discrepancy of high order features between heterogeneous images. Experimental results demonstrate the superiority of the proposed method over existing state-of-the-art methods.
Article
Green & Sustainable Science & Technology
Waqas Ahmad, Hikmat Ullah Khan, Tasswar Iqbal, Muhammad Attique Khan, Usman Tariq, Jae-hyuk Cha
Summary: With the rapid growth of user-generated content on social media, sentiment analysis has become a significant research area. In this manuscript, a technique combining variant algorithms in a parallel manner is proposed to extract advantageous informative features, and then perform sentiment classification. The proposed methodology, a combination of MC-CNN and MC-Bi-GRU, treats them equally in terms of input parameters and shares hidden layer information, making it distinctive and outperforming existing models.
Article
Mathematics
Zihao Lu, Xiaohui Hu, Yun Xue
Summary: This paper proposes a dual-word embedding model considering syntactic information for cross-domain sentiment classification. Experimental results show that the model outperforms other baselines on two real-world datasets.
Article
Computer Science, Artificial Intelligence
Jingyao Tang, Yun Xue, Ziwen Wang, Shaoyang Hu, Tao Gong, Yinong Chen, Haoliang Zhao, Luwei Xiao
Summary: A Bayesian estimation-based sentiment word embedding model has been proposed, which effectively extracts sentiment information of low-frequency words and integrates it into word embedding learning through a novel loss function. Experimental results demonstrate that BESWE outperforms many state-of-the-art methods in sentiment analysis tasks.
CAAI TRANSACTIONS ON INTELLIGENCE TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Rohan Kumar Yadav, Lei Jiao, Morten Goodwin, Ole-Christoffer Granmo
Summary: This paper simplifies preprocessing by incorporating polarity lexicon replacement and masking techniques to eliminate positional embedding. The experimental results demonstrate a significant improvement in accuracy for publicly available ABSA datasets using this approach.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Mathematics, Interdisciplinary Applications
Qi Zhang, Zufan Zhang, Maobin Yang, Lianxiang Zhu
Summary: This paper explores the coevolution of emotional contagion and behavior for microblog sentiment analysis using a deep learning architecture (MSA-UITC). The tie strength between microblogs is used to quantify the relationship between emotional contagion and behavior, leading to the development of a CNN-BiLSTM-Attention network for sentiment analysis. Experimental results show that MSA-UITC outperforms existing state-of-the-art methods on real Twitter datasets.
Article
Computer Science, Information Systems
Yinxue Yi, Zufan Zhang, Laurence T. Yang, Xianjun Deng, Lingzhi Yi, Xiaokang Wang
Summary: The Social Internet of Things integrates social networks and the Internet of Things, facilitating interactions between things, people, and between people and things, resulting in a wealth of information that holds hidden value for social administration and people's lives. Characterizing the interplay between behavior spreading and information diffusion is crucial in predicting and managing information in SIoT. A cloud-edge-aided information diffusion model and a blockchain-based cloud-edge SIoT architecture are proposed for efficient interactions and enhanced security of information diffusion.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Information Systems
Zufan Zhang, Yucheng Yang, Zongming Lv, Chenquan Gan, Qingyi Zhu
Summary: The LMFNet framework, incorporating a deep attentive 3D residual network and a multistage fusion strategy, addresses the limitations of current 3D convolutional networks in human activity recognition, achieving higher recognition accuracy and training efficiency compared to existing methods.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Editorial Material
Mathematics, Interdisciplinary Applications
Chenquan Gan, Qingyi Zhu, Wei Wang, Jianxin Li
DISCRETE DYNAMICS IN NATURE AND SOCIETY
(2021)
Article
Computer Science, Artificial Intelligence
Chenquan Gan, Junhao Xiao, Zhangyi Wang, Zufan Zhang, Qingyi Zhu
Summary: This paper proposes a novel method for facial expression recognition by eliminating redundant information from emotional-unrelated regions, leading to more accurate recognition. The method utilizes a densely connected convolutional neural network with hierarchical spatial attention, and its superior performance is verified through experiments.
IMAGE AND VISION COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Yinxue Yi, Zufan Zhang, Laurence T. Yang, Xiaokang Wang, Chenquan Gan
Summary: This article proposes a control strategy for information diffusion in social Internet of Things (SIoT) based on edge computing and graph neural networks. The strategy can effectively dominate information diffusion within the target scope.
Article
Computer Science, Information Systems
Qingyi Zhu, Pingfan Xiang, Xuhang Luo, Chenquan Gan
Summary: This article proposes a hybrid model to study the propagation of computer viruses on the Internet. Through mathematical analysis and numerical simulations, the equilibria and their stabilities are investigated, and the impact of parameters on system stability is analyzed.
SECURITY AND COMMUNICATION NETWORKS
(2022)
Article
Computer Science, Information Systems
Zufan Zhang, Yue Peng, Chenquan Gan, Andrea Francesco Abate, Lianxiang Zhu
Summary: This paper proposes a lightweight three-dimensional residual attention network (Sep-3D RAN) for human action recognition in video-based human computer interaction. Sep-3D RAN utilizes separable three-dimensional residual attention blocks and a dual attention mechanism to extract more discriminative features and improve model guidance capability. A multi-stage training strategy effectively reduces over-fitting.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Chenquan Gan, Yucheng Yang, Qingyi Zhu, Deepak Kumar Jain, Vitomir Struc
Summary: This paper proposes a hierarchical feature interactive fusion network (DHF-Net) for identifying specific emotions during a dialogue. The network balances contextual information and fine-grained information, and extracts more detailed information. Experimental results demonstrate an improvement in performance.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Mathematics, Applied
Chenquan Gan, Yi Qian, Anqi Liu, Qingyi Zhu
Summary: This paper studies the dynamics of computer virus spread in the Social Internet of Things (SIoT) and proposes a dynamic model that combines the impact of search engines and hierarchical individual-awareness. The theoretical results are fully verified through experiments on real datasets.
COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION
(2022)
Review
Mathematics
Chenquan Gan, Jiabin Lin, Da-Wen Huang, Qingyi Zhu, Liang Tian
Summary: The industrial internet of things (IIoT) integrates traditional industry with modern information technology to improve production efficiency and quality. However, it also faces serious challenges from advanced persistent threats (APTs), a stealthy and persistent method of attack. This paper defines and develops APTs, examines the types of APT attacks in each layer of the IIoT reference architecture, and reviews existing defense techniques. It also models and analyzes APT activities in IIoT to identify their inherent characteristics and patterns, and proposes open research topics and directions for IIoT security.
Article
Computer Science, Information Systems
Xiaoke Li, Zufan Zhang, Chenquan Gan, Yong Xiang
Summary: This study proposes a novel loss function (inter-class difference loss) to address the high statistical error issue in supervised speech emotion recognition systems. An end-to-end network architecture, called response residual network (R-ResNet), is also proposed to improve the efficiency of the speech emotion recognition system. The experimental results demonstrate the advanced performance of the proposed methods.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Mathematics
Qingyi Zhu, Pingfan Xiang, Kefei Cheng, Chenquan Gan, Lu-Xing Yang
Summary: This paper investigates the propagation of network viruses in complex networks, taking into account the combined influence of network topology and hybrid transmission. The study identifies the propagation threshold and explores the impact of vertical transmission on viral spread. It also addresses the problem of dynamically containing the spread with limited resources, utilizing optimal control theory. Simulation experiments verify the effectiveness of the proposed control strategy.
BULLETIN OF THE IRANIAN MATHEMATICAL SOCIETY
(2023)
Article
Computer Science, Artificial Intelligence
Rui Lv, Dingheng Wang, Jiangbin Zheng, Zhao-Xu Yang
Summary: In this paper, the authors investigate tensor decomposition for neural network compression. They analyze the convergence and precision of tensor mapping theory, validate the rationality of tensor mapping and its superiority over traditional tensor approximation based on the Lottery Ticket Hypothesis. They propose an efficient method called 3D-KCPNet to compress 3D convolutional neural networks using the Kronecker canonical polyadic (KCP) tensor decomposition. Experimental results show that 3D-KCPNet achieves higher accuracy compared to the original baseline model and the corresponding tensor approximation model.
Article
Computer Science, Artificial Intelligence
Xiangkun He, Zhongxu Hu, Haohan Yang, Chen Lv
Summary: In this paper, a novel constrained multi-objective reinforcement learning algorithm is proposed for personalized end-to-end robotic control with continuous actions. The approach trains a single model using constraint design and a comprehensive index to achieve optimal policies based on user-specified preferences.
Article
Computer Science, Artificial Intelligence
Zhijian Zhuo, Bilian Chen, Shenbao Yu, Langcai Cao
Summary: In this paper, a novel method called Expansion with Contraction Method for Overlapping Community Detection (ECOCD) is proposed, which utilizes non-negative matrix factorization to obtain disjoint communities and applies expansion and contraction processes to adjust the degree of overlap. ECOCD is applicable to various networks with different properties and achieves high-quality overlapping community detection.
Article
Computer Science, Artificial Intelligence
Yizhe Zhu, Chunhui Zhang, Jialin Gao, Xin Sun, Zihan Rui, Xi Zhou
Summary: In this work, the authors propose a Contrastive Spatio-Temporal Distilling (CSTD) approach to improve the detection of high-compressed deepfake videos. The approach leverages spatial-frequency cues and temporal-contrastive alignment to fully exploit spatiotemporal inconsistency information.
Review
Computer Science, Artificial Intelligence
Laijin Meng, Xinghao Jiang, Tanfeng Sun
Summary: This paper provides a review of coverless steganographic algorithms, including the development process, known contributions, and general issues in image and video algorithms. It also discusses the security of coverless steganography from theoretical analysis to actual investigation for the first time.
Article
Computer Science, Artificial Intelligence
Yajie Bao, Tianwei Xing, Xun Chen
Summary: Visual question answering requires processing multi-modal information and effective reasoning. Neural-symbolic learning is a promising method, but current approaches lack uncertainty handling and can only provide a single answer. To address this, we propose a confidence based neural-symbolic approach that evaluates NN inferences and conducts reasoning based on confidence.
Article
Computer Science, Artificial Intelligence
Anh H. Vo, Bao T. Nguyen
Summary: Interior style classification is an interesting problem with potential applications in both commercial and academic domains. This project proposes a method named ISC-DeIT, which combines data-efficient image transformer architectures and knowledge distillation, to address the interior style classification problem. Experimental results demonstrate a significant improvement in predictive accuracy compared to other state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Shashank Kotyan, Danilo Vasconcellos Vargas
Summary: This article introduces a novel augmentation technique called Dynamic Scanning Augmentation to improve the accuracy and robustness of Vision Transformer (ViT). The technique leverages dynamic input sequences to adaptively focus on different patches, resulting in significant changes in ViT's attention mechanism. Experimental results demonstrate that Dynamic Scanning Augmentation outperforms ViT in terms of both robustness to adversarial attacks and accuracy against natural images.
Article
Computer Science, Artificial Intelligence
Hiba Alqasir, Damien Muselet, Christophe Ducottet
Summary: The article proposes a solution to improve the learning process of a classification network by providing shape priors, reducing the need for annotated data. The solution is tested on cross-domain digit classification tasks and a video surveillance application.
Article
Computer Science, Artificial Intelligence
Dexiu Ma, Mei Liu, Mingsheng Shang
Summary: This paper proposes a method using neural dynamics solvers to solve infinity-norm optimization problems. Two improved solvers are constructed and their effectiveness and superiority are demonstrated through theoretical analysis and simulation experiments.
Article
Computer Science, Artificial Intelligence
Francesco Gregoretti, Giovanni Pezzulo, Domenico Maisto
Summary: Active Inference is a computational framework that uses probabilistic inference and variational free energy minimization to describe perception, planning, and action. cpp-AIF is a header-only C++ library that provides a powerful tool for implementing Active Inference for Partially Observable Markov Decision Processes through multi-core computing. It is cross-platform and improves performance, memory management, and usability compared to existing software.
Article
Computer Science, Artificial Intelligence
Zelin Ying, Dawei Cheng, Cen Chen, Xiang Li, Peng Zhu, Yifeng Luo, Yuqi Liang
Summary: This paper proposes a novel stock market trends prediction framework called SMART, which includes a self-supervised stock technical data sequence embedding model S3E. By training with multiple self-supervised auxiliary tasks, the model encodes stock technical data sequences into embeddings and uses the learned sequence embeddings for predicting stock market trends. Extensive experiments on China A-Shares market and NASDAQ market prove the high effectiveness of our model in stock market trends prediction, and its effectiveness is further validated in real-world applications in a leading financial service provider in China.
Article
Computer Science, Artificial Intelligence
Hao Li, Hao Jiang, Dongsheng Ye, Qiang Wang, Liang Du, Yuanyuan Zeng, Liu Yuan, Yingxue Wang, C. Chen
Summary: DHGAT1, a dynamic hyperbolic graph attention network, utilizes hyperbolic metric properties to embed dynamic graphs. It employs a spatiotemporal self-attention mechanism and weighted node representations, resulting in excellent performance in link prediction tasks.
Article
Computer Science, Artificial Intelligence
Jiehui Huang, Zhenchao Tang, Xuedong He, Jun Zhou, Defeng Zhou, Calvin Yu-Chian Chen
Summary: This study proposes a progressive learning multi-scale feature blending model for image deraining tasks. The model utilizes detail dilation and texture extraction to improve the restoration of rainy images. Experimental results show that the model achieves near state-of-the-art performance in rain removal tasks and exhibits better rain removal realism.
Article
Computer Science, Artificial Intelligence
Lizhi Liu, Zilin Gao, Yinhe Wang, Yongfu Li
Summary: This paper proposes a novel discrete-time interconnected model for depicting complex dynamical networks. The model consists of nodes and edges subsystems, which consider the dynamic characteristic of both nodes and edges. By designing control strategies and coupling modes, the stabilization and synchronization of the network are achieved. Simulation results demonstrate the effectiveness of the proposed methods.
