Article
Business, Finance
Yongan Xu, Chao Liang, Yan Li, Toan L. D. Huynh
Summary: This paper constructs a monthly news-based manager sentiment indicator based on the tone of managers' news reports, showing strong predictability for stock returns, especially in high sentiment periods. For investors, using this forecasting information to optimize stock portfolios can generate significant economic value.
FINANCE RESEARCH LETTERS
(2022)
Article
Computer Science, Artificial Intelligence
Shilpa Gite, Hrituja Khatavkar, Ketan Kotecha, Shilpi Srivastava, Priyam Maheshwari, Neerav Pandey
Summary: The stock market is influenced by complex sentiments and media releases, making price predictions challenging. This paper proposes using machine learning and LSTM to improve accuracy by incorporating sentiment analysis. LSTM has proven effective in learning long-term dependencies, and when combined with historical stock data and news sentiment, it can enhance predictive models.
PEERJ COMPUTER SCIENCE
(2021)
Article
Mathematics
Bledar Fazlija, Pedro Harder
Summary: This paper extracts financial market sentiment information from news articles using natural language processing methods and predicts the price direction of the stock market index. The results show that sentiment scores based on news content are particularly useful for stock price direction prediction.
Article
Mathematics
Marian Pompiliu Cristescu, Raluca Andreea Nerisanu, Dumitru Alexandru Mara, Simona-Vasilica Oprea
Summary: This study explores how sentiment analysis can improve the accuracy of regression models for predicting the opening prices of selected stocks. The findings show that sentiment analysis as an exogenous factor improves the goodness of fit in nonlinear autoregression models, and polynomial autoregressions outperform linear ones.
Article
Management
David Weinbaum, Andrew Fodor, Dmitriy Muravyev, Martijn Cremersd
Summary: The study reveals that the predictiveness of option trading volume on stock prices depends on whether the information is scheduled or unscheduled. Trading costs and margin costs also have an impact on profitability post news announcements.
MANAGEMENT SCIENCE
(2022)
Review
Computer Science, Artificial Intelligence
Shazia Usmani, Jawwad A. Shamsi
Summary: Stock market prediction is a challenging task that requires deep insights for extraction of news events, analysis of historic data, and impact of news events on stock price trends. This paper presents a detailed survey covering key terms and phases of generic stock prediction methodology, challenges, data preprocessing techniques, feature extraction techniques, prediction techniques, and future directions for news sensitive stock prediction. The significance of using structured text features, opinion extraction techniques, domain knowledge, and deep neural network based prediction techniques is highlighted.
PEERJ COMPUTER SCIENCE
(2021)
Article
Computer Science, Information Systems
Ruize Gao, Shaoze Cui, Hongshan Xiao, Weiguo Fan, Hongwu Zhang, Yu Wang
Summary: This study investigates the predictive abilities of different news providers based on sentiment analysis and proposes a framework that assigns different weights to improve prediction performance. The Loughran-McDonald sentiment dictionary is used for sentiment analysis, and the sentiment scores are integrated to obtain the sentiment index of each news provider. Recurrent neural networks are employed to build base classifiers, and the evidential reasoning rule is adopted to combine them for predicting stock market index movement. The genetic algorithm is used to optimize the weights of base classifiers and important hyper-parameters. Experimental results demonstrate the effectiveness of the proposed approach in improving prediction performance, and the trading strategy based on the model's results achieves higher return rates.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Qianyi Xiao, Baha Ihnaini
Summary: Nowadays, the vast amount of data generated on the Internet has become a valuable resource for investors. By using text mining and sentiment analysis techniques, investors can accurately assess confidence in specific stocks to make informed decisions. In this study, two different time divisions were designed to analyze the predictive power of tweets and news on next-day stock trends. The results indicated that the opening hours division outperformed the natural hours division.
PEERJ COMPUTER SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Qing Li, Jinghua Tan, Jun Wang, Hsinchun Chen
Summary: In financial markets, it is believed that market information affects stock movements, creating a multimodal challenge; handling this challenge involves addressing data mode interactions and sampling time heterogeneity; previous research assumed news affects specific stocks, but in reality impacts related stocks.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Cheng Zhao, Xiaohui Liu, Jie Zhou, Yuefeng Cen, Xiaomin Yao
Summary: Most stock price predictive models neglect the correlation effects between stocks, while this article proposes a unified time-series relational multi-factor model that can automatically extract relational features and integrate them with other multiple dimensional features, resulting in significantly improved prediction accuracy and stability.
PEERJ COMPUTER SCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Saleh Albahli, Awais Awan, Tahira Nazir, Aun Irtaza, Ali Alkhalifah, Waleed Albattah
Summary: This paper proposes a deep learning-based framework for predicting stock market using financial news articles. Through preprocessing, feature extraction, feature reduction, and classifier training, the framework achieves an average prediction accuracy of 92.5% on a publicly available dataset, demonstrating its robustness.
COMPLEX & INTELLIGENT SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Xingtong Chen, Xiang Ma, Hua Wang, Xuemei Li, Caiming Zhang
Summary: This research proposes a hierarchical attention network based on attentive multi-view news learning (NMNL) to extract more useful information for stock prediction from news and the stock market. Extensive experiments demonstrate the superiority of NMNL over state-of-the-art stock prediction solutions.
Article
Computer Science, Artificial Intelligence
Wei -Chao Lin, Chih-Fong Tsai, Hsuan Chen
Summary: This study utilized text mining techniques and machine learning algorithms for stock market prediction, finding that the combination of CNN with Word2vec and CNN with BERT performed the best. Additionally, the use of different text feature representations and learning models in financial news articles published on different news platforms can have an impact on prediction results.
APPLIED SOFT COMPUTING
(2022)
Article
Thermodynamics
Yilin Ma, Yudong Wang, Weizhong Wang, Chong Zhang
Summary: This paper explores the use of return and volatility prediction to improve energy portfolio models, using extreme gradient boosting regression trees. Six classical portfolio models are tested, and the results show that using return and volatility prediction significantly enhances these models' performance. Prediction-based weights are also found to be more effective in transforming multiple objectives compared to equal weights. The CVaR-F-PW portfolio performs the best and is recommended for energy stock market portfolio management.
Article
Computer Science, Information Systems
Marwa Sharaf, Ezz El-Din Hemdan, Ayman El-Sayed, Nirmeen A. El-Bahnasawy
Summary: This paper proposes a system that uses sentiment analysis to predict stock market trends during the COVID-19 period, applied to stocks such as TSLA, AMZ, and GOOG. By performing feature selection and reducing spam tweets, the prediction accuracy is improved.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Operations Research & Management Science
Li Zhou, Qinwei Fan, Xiaodi Huang, Yan Liu
Summary: In this paper, a novel variant of the algorithm is proposed to improve the generalization performance for Elman neural networks. By controlling the weight growth and preventing over-fitting, rigorous theoretical analysis and experimental verification have been conducted.
Article
Multidisciplinary Sciences
Qinwei Fan, Zhiwen Zhang, Xiaodi Huang
Summary: This paper presents a novel parametric conjugate gradient method based on the secant equation for training Elman neural network. The theoretical convergence of the algorithm is rigorously proved, and the feasibility and correctness of the method are demonstrated through numerical experiments.
ADVANCED THEORY AND SIMULATIONS
(2022)
Article
Operations Research & Management Science
Yukun Cheng, Xiaotie Deng, Qi Qi, Xiang Yan
Summary: This study examines the problem of resource allocation in a network-based sharing economy, using a pure exchange economy model and applying general equilibrium theory. The focus is on proportional sharing dynamics as a mechanism for network resource sharing, specifically exploring the issue of whether agents may manipulate their private information reports to gain more resources under this mechanism. The study provides the first mathematical proof that this practical distributed network resource-sharing protocol is truthful against manipulative strategies like feasible weight misreporting and edge deletion applied individually and together.
MATHEMATICS OF OPERATIONS RESEARCH
(2023)
Article
Engineering, Biomedical
Yongchun Cao, Liangxia Liu, Xiaoyan Chen, Zhengxing Man, Qiang Lin, Xianwu Zeng, Xiaodi Huang
Summary: In this paper, we propose a deep learning-based segmentation method to automatically identify and delineate metastatic lesions in low-resolution bone scan images. The method utilizes view aggregation and feature extraction to achieve automatic lesion segmentation, and performs well in clinical experiments.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Engineering, Biomedical
Yanru Guo, Qiang Lin, Yubo Wang, Xu Cao, Yongchun Cao, Zhengxing Man, Xianwu Zeng, Xiaodi Huang
Summary: In this study, a deep learning-based image classification model is proposed to improve the accuracy and efficiency of diagnosing lung cancer bone metastasis. The model learns features from two views of an image and aggregates them for classification. Experimental evaluations show that the network performs well in automatically classifying metastatic images.
JOURNAL OF MEDICAL AND BIOLOGICAL ENGINEERING
(2023)
Article
Engineering, Industrial
Shuliang Wang, Xifeng Gu, Jiawei Chen, Chen Chen, Xiaodi Huang
Summary: This paper presents a methodological framework for enhancing the robustness of cyber-physical systems in different failure scenarios. The framework combines an AC power flow model, a routing scheme, and weak interdependencies to accurately characterize and mitigate cascading failures. The proposed improvement strategies have been validated through simulations, demonstrating significant increases in robustness.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Biochemical Research Methods
Weiqi Zhai, Xiaodi Huang, Nan Shen, Shanfeng Zhu
Summary: HPO-based approaches are popular for genomic diagnostics of rare diseases, but they do not fully utilize available information on disease and patient phenotypes. We present a new method called Phen2Disease that prioritizes diseases and genes using semantic similarity between phenotype sets. Our experiments show that Phen2Disease outperforms state-of-the-art methods, especially in cohorts with fewer HPO terms. We also find that patients with higher information content scores have more accurate predictions. Phen2Disease provides ranked diseases and patient HPO terms, offering a novel approach for rare disease diagnostics.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Changqin Huang, Junling Zhang, Xuemei Wu, Yi Wang, Ming Li, Xiaodi Huang
Summary: Multimodal sentiment analysis (MSA) has gained significant attention for incorporating not only text but also audio and visual data. The effective fusion of sentiment information from multiple modalities is crucial for improving MSA performance, but alignment challenges arise during fusion to maintain modality-specific information. This study introduces a Text-centered Fusion Network with crossmodal Attention (TeFNA) that utilizes crossmodal attention to model unaligned multimodal timing information. The proposed TeFNA employs a Text-Centered Aligned fusion method (TCA) that prioritizes the text modality to enhance the representation of fusion features. Additionally, TeFNA maximizes mutual information between modality pairs to preserve task-related emotional information, thereby ensuring the preservation of key modality information from input to fusion. Comprehensive experiments on the CMU-MOSI and CMU-MOSEI multimodal datasets demonstrate that the proposed TeFNA model outperforms existing methods in terms of various metrics.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Xuemei Wu, Jie He, Qionghao Huang, Changqin Huang, Jia Zhu, Xiaodi Huang, Hamido Fujita
Summary: This study proposes a novel cross-hierarchy contrast (CHC) framework called FER-CHC for facial expression recognition tasks. FER-CHC utilizes a contrastive learning mechanism to improve the performance of CNN-based models by utilizing crucial features and globally integrating them through a fusion network. Experimental results demonstrate that FER-CHC achieves state-of-the-art performances on multiple datasets.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Zhenjiao Liu, Zhikui Chen, Yue Li, Liang Zhao, Tao Yang, Reza Farahbakhsh, Noel Crespi, Xiaodi Huang
Summary: This article presents a novel algorithm called IMC-NLT for incomplete multi-view clustering. By utilizing non-negative matrix factorization and a low-rank tensor, IMC-NLT effectively extracts hidden information from incomplete views, overcoming limitations of existing algorithms. Experimental results demonstrate that IMC-NLT outperforms baseline methods and produces stable and promising results.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Xiaodong Li, Chenxin Zou, Pangjing Wu, Qing Li
Summary: The exposure to massive information in daily lives has made it necessary for people to efficiently obtain major points. This article proposes a topic sentiment summarization framework based on reaching definition (TSSRD) to generate high-quality summaries by incorporating sentiment changes and flow. Experimental results demonstrate the effectiveness of the framework.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Qinwei Fan, Fengjiao Zheng, Xiaodi Huang, Dongpo Xu
Summary: This paper proposes a new algorithm for Pi-sigma neural networks with entropy error functions based on L-0 regularization. One of the key features of the proposed algorithm is the use of an entropy error function instead of the more common square error function. The algorithm also employs L-0 regularization to ensure network efficiency. Theoretical analysis and experimental verification prove the monotonicity, strong convergence, and weak convergence of the network. Experiments demonstrate improved performance of the algorithm for classification and regression problems.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2023)
Article
Computer Science, Information Systems
Hongyin Chen, Zhaohua Chen, Yukun Cheng, Xiaotie Deng, Wenhan Huang, Jichen Li, Hongyi Ling, Mengqian Zhang
Summary: This article introduces a reputation-based protocol to help governors evaluate the reliability of collectors, reducing verification workloads by selecting reliable collectors.
IEEE TRANSACTIONS ON CLOUD COMPUTING
(2023)
Article
Engineering, Industrial
Shuliang Wang, Zhaoyang Guo, Xiaodi Huang, Jianhua Zhang
Summary: This paper presents a novel three-stage analysis framework for investigating the resilience of power networks in the face of failures. The framework includes network modeling, resilience metrics, performance analysis, and evaluation. The study shows that degree-based attacks have the biggest impact on reducing network size, while betweenness-based attacks have the fastest decrease in network efficiency. The proposed approach provides valuable insights for decision-makers in developing mitigation techniques and optimal protection strategies.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2024)
Article
Computer Science, Information Systems
Mengqian Zhang, Jichen Li, Zhaohua Chen, Hongyin Chen, Xiaotie Deng
Summary: This article discusses the key issues faced by sharding technology, proposes a committee structure and a reputation mechanism to improve system performance, and presents a recovery process in case of malicious leader behavior.
IEEE TRANSACTIONS ON CLOUD COMPUTING
(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.