Article
Computer Science, Information Systems
Cun Ji, Mingsen Du, Yanxuan Wei, Yupeng Hu, Shijun Liu, Li Pan, Xiangwei Zheng
Summary: Time series classification is widely used in various domains, including EEG/ECG classification, device anomaly detection, and speaker authentication. Despite the existence of many methods, selecting intuitive temporal features for accurate classification remains a challenge. Therefore, this paper proposes a new method called TSC-RTF, which utilizes random temporal features, and shows that it can compete with state-of-the-art methods.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Computer Science, Information Systems
Mingsen Du, Yanxuan Wei, Xiangwei Zheng, Cun Ji
Summary: Multivariate time series classification is widely used in various real-life applications and has attracted significant research interest. However, existing methods only focus on local or global features and overlook the spatial dependency features among multiple variables. In this study, we propose a multi-feature based network (MF-Net) that captures both local and global features through the attention-based mechanism and integrates them to capture spatial dependency features. Experimental results on UEA datasets demonstrate that our method performs competitively with state-of-the-art methods.
INFORMATION SCIENCES
(2023)
Article
Engineering, Electrical & Electronic
Timo De Waele, Adnan Shahid, Daniel Peralta, Anniek Eerdekens, Margot Deruyck, Frank A. M. Tuyttens, Eli De Poorter
Summary: To track the activities and performance of horses, inertial measurement units (IMUs) combined with machine learning algorithms are commonly used. A data-efficient algorithm is proposed that only requires 3 minutes of labeled calibration data. This algorithm achieved a 95% accuracy on datasets captured with leg-mounted IMUs and neck-mounted IMU. However, when the algorithm was calibrated on multiple horses and evaluated on unfamiliar horses, there was a 15% drop in classification accuracy.
IEEE SENSORS JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Bhaskar Dhariyal, Thach Le Nguyen, Georgiana Ifrim
Summary: Accuracy is important in time series classification, but so is speed and data reduction. We propose and evaluate methods for selecting useful channels based on the distance between class prototypes. Our techniques achieve significant data reduction and classifier speedup while maintaining accuracy.
DATA MINING AND KNOWLEDGE DISCOVERY
(2023)
Article
Agronomy
Meng Zhou, Hengbiao Zheng, Can He, Peng Liu, G. Mustafa Awan, Xue Wang, Tao Cheng, Yan Zhu, Weixing Cao, Xia Yao
Summary: This study proposes a classification method based on UAV imagery for crop phenology detection. The results show that the combination of spectral and texture features can improve classification accuracy, providing technical guidance for real-time detection of crop phenology.
FIELD CROPS RESEARCH
(2023)
Article
Computer Science, Interdisciplinary Applications
Shubhranshu Shekhar, Dhivya Eswaran, Bryan Hooi, Jonathan Elmer, Christos Faloutsos, Leman Akoglu
Summary: How can we predict the health outcomes of a cardiac-arrest patient being monitored in the ICU for brain activity as early as possible? Early decision-making is crucial in applications like patient monitoring to enable early intervention and better care. However, predicting outcomes early from EEG data presents challenges such as trade-offs between early predictions and accuracy, processing large-scale and streaming data, and handling multi-variate and multi-length time series. To address this, BeneFitter integrates early prediction savings and misclassification costs into a unified target called benefit, providing efficient real-time decision-making, handling diverse patient data, and achieving time-savings with comparable or better accuracy than competitors.
JOURNAL OF BIOMEDICAL INFORMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Weibo Shu, Yaqiang Yao, Shengfei Lyu, Jinlong Li, Huanhuan Chen
Summary: In the research area of time series classification, a novel algorithm called short isometric shapelet transform (SIST) is introduced in this paper to reduce time complexity by fixing the length of shapelet and training a single linear classifier. The theoretical evidence and empirical experiments demonstrate the superior performance of the proposed algorithm in terms of near-lossless accuracy while reducing time complexity.
KNOWLEDGE AND INFORMATION SYSTEMS
(2021)
Article
Environmental Sciences
Jin Yan, Yuanyuan Chen, Jiazhu Zheng, Lin Guo, Siqi Zheng, Rongchun Zhang
Summary: This research utilizes advanced remote sensing technology and multi-source time series images to accurately extract and identify urban forests. The results show that the improved feature selection method has a high classification accuracy and is important for the management and monitoring of urban forests.
Article
Computer Science, Artificial Intelligence
Guiling Li, Shaolin Xu, Senzhang Wang, Philip S. Yu
Summary: Time series classification is an important task in time series data mining. This paper proposes a new TSC algorithm called FIT, which combines appropriate transformation series and interval features, and adaptsively converts the interval features of each series during formal training. Experimental results on 85 UCR time series classification datasets show that FIT achieves better accuracy while maintaining high efficiency compared to state-of-the-art methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Environmental Sciences
Peng Wei, Huichun Ye, Shuting Qiao, Ronghao Liu, Chaojia Nie, Bingrui Zhang, Lijuan Song, Shanyu Huang
Summary: Early-season crop mapping and information extraction are crucial for crop growth monitoring and yield prediction. This study demonstrates that the random forest importance analysis can select suitable features for early crop classification, and remote sensing can extract crop acreage information within the reproductive period.
Article
Computer Science, Artificial Intelligence
Hegui Zhu, Jiapeng Zhang, Hao Cui, Kai Wang, Qingsong Tang
Summary: Currently, the most popular and effective approach to solve multivariate time series classification tasks is based on deep learning technology. However, existing deep learning-based algorithms ignore the unique time characteristics of time series and the correlation between features in different convolutional layers, resulting in unsatisfactory classification accuracy. To address this issue, we propose a new time corrected residual attention network (TCRAN) that can extract long-term time-dependence information to enhance the discriminative power of the network. Experimental results show that TCRAN achieves the highest average classification accuracy of 0.7276 and improves accuracy by 1.64% compared to the state-of-the-art method.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Cun Ji, Mingsen Du, Yupeng Hu, Shijun Liu, Li Pan, Xiangwei Zheng
Summary: With the increasing application of Internet of Things technology, time series classification has become a research hotspot in the field of data mining. This paper proposes a new method for time series classification based on temporal features (TSC-TF), which generates temporal feature candidates through time series segmentation and selects important features with the help of a random forest. The experimental results on various datasets demonstrate the superiority of the proposed method.
APPLIED SOFT COMPUTING
(2022)
Article
Environmental Sciences
Dino Dobrinic, Mateo Gasparovic, Damir Medak
Summary: This study assessed the classification accuracy of land-cover (LC) mapping using Sentinel satellite data, particularly focusing on vegetation classes. By combining Sentinel-1 (S1) and Sentinel-2 (S2) data with the Random Forest (RF) method, a high overall accuracy of 91.78% was achieved.
Article
Computer Science, Artificial Intelligence
Ziheng Duan, Haoyan Xu, Yueyang Wang, Yida Huang, Anni Ren, Zhongbin Xu, Yizhou Sun, Wei Wang
Summary: Multivariate time-series classification (MTSC) has gained considerable attention in recent years. Existing deep-learning-based techniques focus on the temporal dependency of a single time series. This study proposes a novel graph-pooling-based framework, MTPool, to address the limitations of existing methods. MTPool combines graph neural networks and variational graph pooling to achieve global graph representation learning and graph coarsening.
Article
Computer Science, Information Systems
Cun Ji, Yupeng Hu, Shijun Liu, Li Pan, Bo Li, Xiangwei Zheng
Summary: This study combines a fully convolutional network with shapelet features to address the low efficiency and inadequate accuracy of shapelet feature extraction in time series classification. Experimental results demonstrate that the proposed method achieves high accuracy and more effectively extracts shapelet features.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Yiqi Chen, Tieyun Qian
INFORMATION SCIENCES
(2020)
Article
Information Science & Library Science
Xuhui Li, Liuyan Liu, Xiaoguang Wang, Yiwen Li, Qingfeng Wu, Tieyun Qian
Summary: The paper proposes a graph-based representation approach for evolutionary knowledge under big data circumstances, introducing a semantic data model called MGraph. The MGraph utilizes directed acyclic graph-like types to specify the structural features of knowledge with intention variety and proposes several specialization mechanisms for knowledge evolution. This approach effectively addresses the major issues of evolutionary knowledge from big data and is promising in building a big knowledge base.
ELECTRONIC LIBRARY
(2021)
Article
Mathematics, Interdisciplinary Applications
Kejian Tang, Shaohui Zhan, Tao Zhan, Hui Zhu, Qian Zeng, Ming Zhong, Xiaoyu Zhu, Yuanyuan Zhu, Jianxin Li, Tieyun Qian
Summary: When promoting a business or activity in geo-social networks, the Distance-Aware Influence Maximization (DAIM) problem becomes crucial. Two efficient location sampling approaches, FLS and CFLS, were proposed and validated on real datasets, demonstrating their effectiveness in solving the DAIM problem.
Article
Computer Science, Information Systems
Ke Sun, Tieyun Qian, Xu Chen, Ming Zhong
Summary: In this paper, we propose a context-aware seq2seq translation model to capture the inter-sequence dependency for sequential recommendations. The injected VAE in our model redresses the semantic imbalance between context and item, leading to superior performance over state-of-the-art baselines.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Zhuang Chen, Tieyun Qian
Summary: This paper proposes a novel approach for sequence tagging data augmentation, which uses descriptions to supervise instance-level augmentation process in order to consistently generate high-quality synthetic data; and retrieves demonstrations to enhance the learning capability of neural networks under limited training data.
WORLD WIDE WEB-INTERNET AND WEB INFORMATION SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Wanli Li, Tieyun Qian, Ming Zhong, Xu Chen
Summary: This article presents a novel semisupervised relation extraction (RE) method that requires only small human efforts and is robust to the size of the initial labeled data. By constructing lexical and semantic graphs and utilizing graph interaction, the method successfully recognizes high-quality unlabeled samples.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Acoustics
Zhuang Chen, Tieyun Qian
Summary: This study proposes a novel domain adaptation method to enhance the transferability of source words, achieving cross-domain aspect term extraction and aspect-level sentiment classification. Experimental results show that our method significantly outperforms state-of-the-art methods by introducing transferable prototypes, with an absolute F1 increase of 3.95% over the best baseline.
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Tieyun Qian, Yile Liang, Qing Li, Hui Xiong
Summary: Rating prediction is a classic problem addressed by matrix factorization, but recent advancements in deep learning, particularly graph neural networks, have shown impressive progress. This study introduces a new AGNN framework that utilizes attribute graphs to learn preference embeddings for strict cold start users/items.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Automation & Control Systems
Jia Chen, Ming Zhong, Jianxin Li, Dianhui Wang, Tieyun Qian, Hang Tu
Summary: This article focuses on the "oversmoothing" problem in attributed network representation learning, proposing to evaluate a smoothing parameter based on network topological characteristics to adaptively smooth node attributes and structure information, resulting in robust and distinguishable node features.Extensive experiments show that this approach effectively preserves the intrinsic information of networks compared to state-of-the-art works on benchmark datasets with varying topological characteristics.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Computer Science, Artificial Intelligence
Xiaoying Chen, Mi Zhang, Shengwu Xiong, Tieyun Qian
Summary: This study proposes a new method that utilizes the sequential form of POS tags in sentences to bridge the gap between the original sentence and imperfect parse tree, while enabling the learned POS embeddings to correspond and interact with word embeddings, achieving significant advantages in relation extraction tasks.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Tieyun Qian, Yile Liang, Qing Li, Xuan Ma, Ke Sun, Zhiyong Peng
Summary: The recommendation of tail items is crucial in recommender systems. Existing methods often overlook the importance of tail items and suffer from a decline in accuracy. This work proposes a unified framework to optimize accuracy and novelty targets, and introduces a novel approach for the recommendation of cold-start items.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Wanli Li, Tieyun Qian, Xuhui Li, Lixin Zou
Summary: To address the shortage of labeled data for relation extraction tasks, a novel adversarial multi-teacher distillation (AMTD) framework is proposed, which improves the performance of semi-supervised relation extraction methods through multi-teacher knowledge distillation and adversarial training.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Acoustics
Mi Zhang, Tieyun Qian, Bing Liu
Summary: In this paper, a multi-scale representation and metric learning framework is proposed for relation extraction tasks. By constructing hierarchy of lexical and syntactic features and relations, the framework utilizes multi-scale convolutional neural network and graph convolutional network for feature aggregation and expanding receptive field, as well as multi-scale metric learning to exploit relations between features and samples.
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
(2022)
Proceedings Paper
Computer Science, Information Systems
Yile Liang, Tieyun Qian, Qing Li, Hongzhi Yin
Summary: Recommender systems are crucial in online platforms, offering personalized recommendations. This study focuses on the challenge of balancing accuracy and diversity, aiming to enhance domain-level and user-level adaptivity.
SIGIR '21 - PROCEEDINGS OF THE 44TH INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhuang Chen, Tieyun Qian
Summary: This study introduces a novel active domain adaptation method that significantly outperforms previous approaches by constructing syntactic and semantic bridges to transfer aspect terms.
59TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS AND THE 11TH INTERNATIONAL JOINT CONFERENCE ON NATURAL LANGUAGE PROCESSING, VOL 1 (ACL-IJCNLP 2021)
(2021)
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.