Article
Computer Science, Artificial Intelligence
Qiuyu Kong, Jin Tang, Chenglong Li, Xin Wang, Jian Zhang
Summary: The study demonstrates that utilizing the complementary properties of different CNNs can improve visual tracking performance. By jointly inferring candidate location, predicted location, and confidence score, the importance of different CNNs is identified, and the adaptive fusion of prediction scores enhances tracking robustness.
COGNITIVE COMPUTATION
(2022)
Article
Environmental Sciences
Qing Liu, Yongsheng Dong, Zhiqiang Jiang, Yuanhua Pei, Boshi Zheng, Lintao Zheng, Zhumu Fu
Summary: With the development of image segmentation technology, the importance of image context information in semantic segmentation has been recognized. In order to capture rich context information effectively, we proposed a Multi-Pooling Context Network (MPCNet) for image semantic segmentation. The network includes Pooling Context Aggregation Module and Spatial Context Module to capture deep context information and detailed spatial context respectively. Experimental results on multiple datasets demonstrate the effectiveness of our proposed network in context extraction.
Article
Computer Science, Artificial Intelligence
Ning Tong, Ying Tang, Bo Chen, Lirong Xiong
Summary: A semi-supervised representation learning model, RANCH, is proposed using a graph attention network and a convolutional neural network for heterogeneous information networks. Experimental results show that the model outperforms state-of-the-arts in node classification on three real-world datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Liancheng He, Liang Bai, Xian Yang, Hangyuan Du, Jiye Liang
Summary: Traditional GCNs have the over-smoothing problem, limiting their ability to extract high-order information and obtain robust data representation. To address this issue, we propose a novel high-order graph attention network that adaptively aggregates node features from multi-hop neighbors through an attention mechanism. We also update the graph by adjusting the edges with small step sizes using the aggregated node representation. Theoretical analysis demonstrates the relationships between our proposed model and other GCN models, and experimental results show the superiority of our proposed model over other models.
INFORMATION SCIENCES
(2023)
Article
Agriculture, Multidisciplinary
Xue Zhao, Kaiyu Li, Yunxia Li, Juncheng Ma, Lingxian Zhang
Summary: This study developed a new vegetable disease identification model, DTL-SE-ResNet50, and compared it with other models. The results showed that the new model had high identification precision and fast identification speed.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Microbiology
Tsi-Shu Huang, Kevin Wang, Xiu-Yuan Ye, Chii-Shiang Chen, Fu-Chuen Chang
Summary: This study utilizes transfer learning with CNNs to classify fungal genera and identify Aspergillus species using microscopic images. With the involvement of medical technologists, the study achieved high classification accuracy and highlights the potential of merging advanced technology with medical laboratory practices.
MICROBIOLOGY SPECTRUM
(2023)
Article
Chemistry, Medicinal
Jack Scantlebury, Lucy Vost, Anna Carbery, Thomas E. Hadfield, Oliver M. Turnbull, Nathan Brown, Vijil Chenthamarakshan, Payel Das, Harold Grosjean, Frank von Delft, Charlotte M. Deane
Summary: In recent years, multiple machine learning-based scoring functions have been developed to predict the binding of small molecules to proteins. These scoring functions aim to approximate the distribution that takes two molecules as input and outputs the energy of their interaction. However, many scoring functions rely on data set biases rather than understanding the physics of binding. To test the learning ability of machine learning-based scoring functions, input attribution can be applied to identify important binding interactions. A machine learning-based scoring function was built, which achieved comparable performance to other methods on benchmark tests. Attribution was then used to extract important binding pharmacophores and improve docking scores compared to traditional approaches.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Computer Science, Artificial Intelligence
Sujit Kumar Das, Suyel Namasudra, Awnish Kumar, Nageswara Rao Moparthi
Summary: This paper presents an efficient approach based on Convolutional Neural Network (CNN) called AESPNet for the identification of Diabetic Foot Ulcer (DFU). Compared with other standard CNN-based schemes, AESPNet demonstrates better performance in DFU classification.
IMAGE AND VISION COMPUTING
(2023)
Article
Computer Science, Software Engineering
Qian Luo, Jie Shao, Wanli Dang, Long Geng, Huaiyu Zheng, Chang Liu
Summary: In this work, we propose an efficient multi-scale channel attention network (EMCA) to address the challenges of occlusion and similar appearance in person re-identification. The EMCA incorporates a novel cross-channel attention module (CCAM) that includes local cross-channel interaction (LCI) and channel weight integration (CWI). Experimental results on popular person re-identification datasets demonstrate that our EMCA outperforms existing state-of-the-art methods consistently.
Article
Computer Science, Information Systems
Yi-Xiang Luo, Jiann-Liang Chen
Summary: This study introduces a Dual Attention Forgery Detection Network that embeds two attention mechanisms to identify traces of tampering in fake videos. The proposed DAFDN outperforms other methods in two benchmark datasets, DFDC and FaceForensics++.
Article
Engineering, Electrical & Electronic
Xinxin Shan, Yutao Shen, Haibin Cai, Ying Wen
Summary: In this paper, a novel network optimization module called CRA is proposed, which enhances the representational power of networks by utilizing the spatial information of feature maps and channel attention. Experiments demonstrate that embedding the CRA module into various networks effectively improves the performance under different evaluation standards.
DIGITAL SIGNAL PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Qing Liu, Yongsheng Dong, Yuanhua Pei, Lintao Zheng, Lei Zhang
Summary: In this paper, a Long and Short-Range Relevance Context Network is proposed to capture global semantic context and local spatial context information. The network utilizes Long-Range Relevance Context Module and Short-Range Relevance Context Module to improve the accuracy of pixel classification and detailed pixel location. A coding and decoding structure is adopted to enhance the segmentation results, and experiments on multiple datasets validate the effectiveness of the network.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Agronomy
Jianwu Lin, Xiaoyulong Chen, Renyong Pan, Tengbao Cao, Jitong Cai, Yang Chen, Xishun Peng, Tomislav Cernava, Xin Zhang
Summary: This study proposes a lightweight CNN model called GrapeNet for the identification of different symptom stages for specific grape diseases. The model achieves excellent classification performance and reduces the number of parameters compared to other classical models.
Article
Chemistry, Analytical
Baosheng Wang, An Lu, Ling Yu
Summary: In this study, a deep learning method combined with a hyperspectral imaging system was developed for quality-based identification of rice samples from different origins. By focusing on the deep features of spectral information, the spectral characteristics of rice samples from different origins were effectively identified. This provides an effective technical method for tracing rice.
ANALYTICAL METHODS
(2023)
Article
Engineering, Industrial
Suckwon Hong, Juram Kim, Han-Gyun Woo, Young-Choon Kim, Changyong Lee
Summary: This study proposes an analytical framework for screening technological ideas in the early stages of development. By associating technical descriptions with patent forward citations, the framework can assess the value of ideas using only the technical content. The framework, applied in the field of pharmaceutical technology, outperforms existing models in terms of accuracy and reliability.
Article
Computer Science, Artificial Intelligence
Qiang Liu, Shu Wu, Liang Wang
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2017)
Article
Computer Science, Artificial Intelligence
Qiang Liu, Feng Yu, Shu Wu, Liang Wang
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY
(2018)
Article
Computer Science, Artificial Intelligence
Qiang Cui, Shu Wu, Qiang Liu, Wen Zhong, Liang Wang
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2020)
Article
Computer Science, Artificial Intelligence
Zeyu Cui, Feng Yu, Shu Wu, Qiang Liu, Liang Wang
Summary: Item representations in recommendation systems are traditionally done using single latent vectors, but utilizing attribute information has recently become popular for better item representations. This article proposes a fine-grained Disentangled Item Representation (DIR) method, representing items as separate attribute vectors for more detailed item information. Experimental results using the LearnDIR strategy show that models developed under DIR framework are effective and efficient, even outperforming state-of-the-art methods in cold-start situations.
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
Jinghao Zhang, Yanqiao Zhu, Qiang Liu, Mengqi Zhang, Shu Wu, Liang Wang
Summary: Multimedia recommendation has become a popular topic in recent years due to the prevalence of multimedia contents on the modern Web. However, previous studies have limitations in modeling item relationships and fusing multiple modalities effectively. To address these issues, this study proposes the MICRO model, which includes a modality-aware structure learning module and a multimodal contrastive framework. Experimental results on real-world datasets demonstrate the superiority of the MICRO model over existing methods.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Yabo Chu, Enneng Yang, Qiang Liu, Yuting Liu, Linying Jiang, Guibing Guo
Summary: Multi-behavior recommendation utilizes auxiliary behaviors to improve the prediction for target behaviors. However, the assumption that all auxiliary behaviors are positively correlated with target behaviors may not hold in real-world datasets. In this paper, we propose a Bi-directional Contrastive Distillation (BCD) model to distill valuable knowledge from the interplay of multiple user behaviors. Experimental results show that our approach outperforms other counterparts in accuracy.
MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2022, PT I
(2023)
Article
Computer Science, Information Systems
Kashan Ahmed, Syed Khaldoon Khurshid, Sadaf Hina
Summary: This paper mainly introduces the construction of the cyber threat intelligence knowledge graph and the information extraction technique. By using joint extraction technique, it solves the problem of traditional techniques becoming ineffective due to the increasing size of CTI data. Experimental results show that this technique outperforms state-of-the-art models in knowledge triple extraction on CTI data and improves the F1 score.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Xinlong He, Yang Xu, Sicong Zhang, Weida Xu, Jiale Yan
Summary: This paper proposes a new membership inference attack method in federated learning, which utilizes data poisoning and sequence prediction confidence. The attack is effective and results in minimal overall model performance degradation.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Tieming Chen, Huan Zeng, Mingqi Lv, Tiantian Zhu
Summary: In this paper, the authors propose a deep learning based dynamic malware detection method called CTIMD, which integrates threat knowledge from CTIs into the learning process of API call sequences with runtime parameters. Experimental results show that CTIMD outperforms existing methods in terms of performance.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Wonwoo Choi, Minjae Seo, Seongman Lee, Brent Byunghoon Kang
Summary: This paper proposes SUM, a backward-edge control flow protection scheme for ARM Cortex-M processors. It combines MPU and the overlooked hardware feature FaultMask to achieve efficient and robust protection. The empirical evaluation shows minimal runtime overhead for the proposed solution.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Liliana Ribeiro, Ines Sousa Guedes, Carla Sofia Cardoso
Summary: Phishing susceptibility is influenced by individual and contextual factors. The study found that individuals who perceive themselves as capable of detecting phishing and those who use online services more frequently are more susceptible to phishing. However, technology competencies and other individual variables do not predict phishing susceptibility.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Wenjie Wang, Yuanhai Shao, Yiju Wang
Summary: In this paper, we investigate the adversarial perturbations of twin support vector machines (TWSVMs) and propose an optimization framework, which provides explicit solutions to increase the interpretability of the conclusion and convenience for calculation.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Snofy D. Dunston, V. Mary Anita Rajam
Summary: This paper proposes a novel adversarial attack technique that can synthesize adversarial images to mislead deep learning models, and also studies interpretability plots. The research findings show that the proposed attack technique influences the interpretability plots, regardless of the success of the attack.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Junchen Li, Guang Cheng, Zongyao Chen, Peng Zhao
Summary: Protocol Reverse Engineering (PRE) is a direct approach for analyzing unknown traffic. This paper proposes a method for clustering unknown traffic based on private protocol labels, and the experimental results demonstrate its advantages on real-world network traffic.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Rafal Kozik, Massimo Ficco, Aleksandra Pawlicka, Marek Pawlicki, Francesco Palmieri, Michal Choras
Summary: The inclusion of Explainability of Artificial Intelligence (xAI) has become a mandatory requirement for designing and implementing reliable, interpretable, and ethical AI solutions. However, it has been shown that xAI can enable successful adversarial attacks in the domain of fake news detection, leading to a decrease in AI security. This paper presents an attack scheme that uses an explainable solution to reshape the structure of the original message, allowing the adversary to manipulate the model's prediction while keeping the message's meaning intact.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Benyuan Yang, Lili Luo, Zhimeng Wang
Summary: Interoperation is widely used in practical industrial applications, but merging local access control policies may lead to security violations. Dealing with these issues in a multidomain environment is critical, but finding the maximum secure interoperation among individual systems poses a challenge due to the large number of entities and access involved.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Binghui Zou, Chunjie Cao, Longjuan Wang, Sizheng Fu, Tonghua Qiao, Jingzhang Sun
Summary: The ongoing struggle between security researchers and malware has led to the exploration of using convolutional neural networks and capsule networks for classification and identification of malware. However, training these networks requires a significant amount of data and parameters, and the research on capsule networks is still in its early stages, posing challenges.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Hongsong Chen, Xingyu Li, Wenmao Liu
Summary: Multivariate time-series anomaly detection is crucial for maintaining normal operation of physical equipment. Recent advances have been made in this field, but two challenges have limited the model's ability to generalize. To address these challenges, a multivariate time-series anomaly detection model consisting of a characterization network and a forecasting network is proposed. Experimental results demonstrate that this method outperforms baseline methods in terms of detection performance and robustness.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Roberto Doriguzzi-Corin, Domenico Siracusa
Summary: This paper discusses the application of federated learning in the field of cybersecurity and proposes an adaptive mechanism-based federated learning solution for DDoS attack detection in dynamic cybersecurity scenarios. Through experiments, it is demonstrated that the proposed solution outperforms state-of-the-art federated learning algorithms in terms of convergence time and accuracy.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Antonio Giovanni Schiavone
Summary: The usage of HTTPS protocol is crucial for secure communication with websites, ensuring the confidentiality, integrity, and authenticity of online data transmissions. The Municipality2HTTPS research project analyzed the implementation of HTTPS in Italian municipalities' websites and identified areas for improvement.
COMPUTERS & SECURITY
(2024)
Article
Computer Science, Information Systems
Domna Bilika, Nikoletta Michopoulou, Efthimios Alepis, Constantinos Patsakis
Summary: Voice Assistants (VAs) are widely used in smart devices, but are vulnerable to attacks, as shown by experiments with popular VAs revealing successful attack rates exceeding 30% and statistical variations among vendors, calling for additional countermeasures to protect user information.
COMPUTERS & SECURITY
(2024)