Article
Computer Science, Information Systems
Vasilis Vouvoutsis, Fran Casino, Constantinos Patsakis
Summary: Malware authors constantly improve their code to evade analysis, making detection difficult. This research proposes complementing sandbox execution with binary emulation frameworks, achieving high accuracy and low computational overhead.
JOURNAL OF INFORMATION SECURITY AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Mamoru Mimura
Summary: There has been a significant increase in malware attacks and malicious traffic. Various machine learning-based detection models have been developed, but their evaluation methods and datasets differ, making it difficult to compare their performances accurately. This study proposes a new metric for evaluating accuracy degradation caused by increasing the benign sample size in binary classification. Using the FFRI dataset, the classification accuracy was evaluated with extracted strings from malware, and it was found that increasing the benign sample size resulted in a decrease in the F1 score.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Alessandro Cucchiarelli, Christian Morbidoni, Luca Spalazzi, Marco Baldi
Summary: This paper presents a methodology for detecting DGA generated domain names using supervised machine learning, which achieves good accuracy and effectiveness in classifying previously unseen domains. The approach is based on lexical features and outperforms some state-of-the-art featureless classification methods based on deep learning.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Daniel Gibert, Carles Mateu, Jordi Planes, Joao Marques-Silva
Summary: Malicious software poses a serious threat on the internet, with traditional detection methods struggling to keep up. Machine learning and deep learning engines have shown promise in handling complex malware and new variants effectively. Further research is needed to improve classification performance and vulnerabilities to adversarial examples.
COMPUTERS & SECURITY
(2021)
Article
Computer Science, Artificial Intelligence
Muhammad Irfan Yousuf, Izza Anwer, Ayesha Riasat, Khawaja Tahir Zia, Suhyun Kim
Summary: The researchers propose a static malware detection system that can detect Portable Executable (PE) malware in Windows environment with high accuracy. By collecting malware samples and extracting relevant information, they combine machine learning, ensemble learning, and dimensionality reduction techniques to construct a system with a detection rate of 99.5% and an error rate of only 0.47%.
PEERJ COMPUTER SCIENCE
(2023)
Review
Computer Science, Artificial Intelligence
Rosmalissa Jusoh, Ahmad Firdaus, Shahid Anwar, Mohd Zamri Osman, Mohd Faaizie Darmawan, Mohd Faizal Ab Razak
Summary: Android is a free open-source operating system widely used by manufacturers to produce mobile devices, but unethical authors often develop malware for various purposes. While practitioners conduct intrusion detection analyses like static analysis, there is a lack of review articles discussing research efforts in this area.
PEERJ COMPUTER SCIENCE
(2021)
Article
Computer Science, Information Systems
Alejandro Guerra-Manzanares, Hayretdin Bahsi
Summary: This study evaluates the machine learning models in the Android malware detection domain and highlights the significance of timestamp selection in concept drift modeling and long-term performance of the model.
COMPUTERS & SECURITY
(2022)
Article
Optics
Alessandro Lupo, Lorenz Butschek, Serge Massar
Summary: The optical domain is a promising field for implementing neural networks due to its speed and parallelism. Extreme Learning Machines (ELMs) are feed-forward neural networks where only output weights are trained, while internal connections remain untrained. Experimental results show that photonic ELM performs well in classification tasks and nonlinear channel equalization tasks.
Article
Computer Science, Information Systems
Fan Ou, Jian Xu
Summary: This paper proposes a novel static sensitive subgraph-based feature for Android malware detection, named S(3)Feature. By mining sensitive subgraphs and neighbor subgraphs to characterize suspicious behaviors of applications, and encoding them into feature vectors, S(3)Feature achieves a 97.04% F1 score in malware detection, outperforming other well-studied features.
COMPUTERS & SECURITY
(2022)
Article
Multidisciplinary Sciences
German Rios-Toledo, Juan Pablo Francisco Posadas-Duran, Grigori Sidorov, Noe Alejandro Castro-Sanchez
Summary: The analysis of an author's writing style involves the characterization and identification of linguistic features. This can be done through extrinsic or intrinsic analysis, with n-grams being a commonly used style marker. In this study, characters, words, Part-Of-Speech (POS) tags, and syntactic relations n-grams were used to analyze novels by eleven English-speaking authors. Results showed significant changes in writing style over time for all authors, with syntactic relations n-grams performing competitively.
Article
Computer Science, Information Systems
Mo'ath Zyout, Raed Shatnawi, Hassan Najadat
Summary: With the advancement of smartphone technology, the development of mobile applications is rapidly growing. These apps are vulnerable to malicious user attacks and differentiating between benign and malicious malware applications is a challenge. This paper proposes two methods, Conv1d and LSTM, for classifying mobile applications into benign or malicious using binary and text encoding techniques. The results show that Conv1d with binary classification outperforms the LSTM model when compared with the Mal-Prem dataset.
INTERNATIONAL JOURNAL OF INFORMATION SECURITY
(2023)
Article
Telecommunications
Jun Liu, Kai Mei, Xiaochen Zhang, Des McLernon, Dongtang Ma, Jibo Wei, Syed Ali Raza Zaidi
Summary: This paper proposes a novel synchronization scheme based on extreme learning machine (ELM) for achieving high-precision synchronization in multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) systems. Simulation results demonstrate that the proposed ELM-based scheme outperforms the traditional method under both additive white Gaussian noise and frequency selective fading channels, without requiring perfect channel state information (CSI) and excessive computational complexity.
IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING
(2022)
Article
Environmental Sciences
Hadi Kardhana, Jonathan Raditya Valerian, Faizal Immaddudin Wira Rohmat, Muhammad Syahril Badri Kusuma
Summary: Jakarta is facing recurring floods due to population growth and the development of slums in vulnerable areas. Using satellite remote sensing data and neural networks can predict water levels, providing more lead time for flood preparedness.
Review
Chemistry, Multidisciplinary
Nana Kwame Gyamfi, Nikolaj Goranin, Dainius Ceponis, Antanas Cenys
Summary: This article provides an analysis of the current state of machine learning techniques for malware detection. It discusses methods for feature extraction and selection to improve the accuracy and precision of detection algorithms. The article also compares and analyzes different machine learning approaches, highlighting their strengths, weaknesses, and performance in detecting system-level malware. It concludes with future research opportunities and serves as a resource for researchers and cybersecurity professionals interested in advancing automated system-level malware detection using machine learning.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Yun Gao, Hirokazu Hasegawa, Yukiko Yamaguchi, Hajime Shimada
Summary: This study proposes a malware classification system based on Control-Flow Graph (CFG) and Graph Isomorphism Network (GIN) using machine learning methods to process large-scale data. Experimental results show that the method achieves high accuracy and AUC in malware detection.
Article
Engineering, Electrical & Electronic
Ahmed Musleh, Guo Chen, Zhao Yang Dong, Chen Wang, Shiping Chen
Summary: Two FDIA characterization algorithms based on PCA and CCA are developed in this paper, with testing results indicating promising performance in FDIA characterization utilizing both algorithms.
IEEE TRANSACTIONS ON POWER SYSTEMS
(2022)
Article
Computer Science, Information Systems
Vijay Varadharajan, Kallol Krishna Karmakar, Uday Tupakula, Michael Hitchens
Summary: This paper addresses the fundamental issue of trust in network slices and proposes a trust model and property-based trust attestation mechanisms. The model evaluates the trust of virtual network functions in the network slice and helps determine their trustworthiness and required properties. The proposed trust model and mechanisms enable service providers to determine the trustworthiness of network services and users to develop trustworthy applications.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2022)
Article
Computer Science, Information Systems
Qin Wang, Longxia Huang, Shiping Chen, Yang Xiang
Summary: In this article, a framework called pAuditChain is proposed to address the data auditing and privacy protection issues of smart meter bills using homomorphic encryption and blockchain technology. The framework not only accommodates individual consumption checking requests but also handles bulk auditing requests from governments. The proposed solution improves the security and privacy of bills without compromising the auditing function.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Theory & Methods
Qin Wang, Jiangshan Yu, Shiping Chen, Yang Xiang
Summary: Limitations in latency and scalability of classical blockchain systems hinder their adoption and application. Reconstructed blockchain systems using Directed Acyclic Graph (DAG) have been proposed to address these limitations and enable fast confirmation and high scalability. However, there is a need for systematic work that summarizes DAG techniques in this field. This Systematization of Knowledge (SoK) provides a comprehensive analysis of existing and ongoing DAG-based blockchain systems, evaluating them from various perspectives and discussing trade-offs, challenges, and future research directions.
ACM COMPUTING SURVEYS
(2023)
Article
Automation & Control Systems
Ahmed S. Musleh, Guo Chen, Zhao Yang Dong, Chen Wang, Shiping Chen
Summary: Automatic generation control is crucial for power grid stability, but its dependence on communication systems makes it vulnerable to cyberphysical attacks. This article proposes a novel spatio-temporal learning algorithm to address the issue of false data injection attacks by learning the normal dynamics and evaluating reconstruction residuals for improved security.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Engineering, Electrical & Electronic
Ahmed S. Musleh, Guo Chen, Zhao Yang Dong, Chen Wang, Shiping Chen
Summary: The utilization of distributed generation units has increased the complexity of power distribution systems. To address this issue, a spatio-temporal learning algorithm is proposed to detect false data injection attacks by assessing the residual error of measurement samples. This data-driven method overcomes the nonlinearities and uncertainties of distribution systems.
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2023)
Article
Chemistry, Multidisciplinary
Kaimeng Ding, Shiping Chen, Yue Zeng, Yingying Wang, Xinyun Yan
Summary: The proposed Transformer-based subject-sensitive hashing algorithm in this paper can be applied to the data security of HRRS images by providing integrity authentication services and generating digital watermarks. It overcomes the shortcomings of existing authentication methods and achieves subject-sensitive authentication of HRRS images. The algorithm enhances robustness, especially against JPEG compression, compared to existing algorithms.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Cybernetics
Geng Sun, Wei Wei, Tingru Cui, Dongming Xu, Shiping Chen, Alex Shvonski, Li Li, Jun Shen, Soheila Garshasbi
Summary: Since the outbreak of COVID-19, there has been a high demand for alternative methods of remote learning to keep students on track and prevent them from being exposed to the risk of infection. Education providers have been experimenting with delivering knowledge and learning materials remotely, combining learning management systems, open educational resources, mini applications in social media, and video-conference software to create multi-channel delivery modes. However, the lack of learner information and the continuous release of new resources have posed challenges in implementing innovative and adaptive micro learning. To address the data sparsity issue, an online computation method has been proposed, along with a lightweight learner-micro-OER profile and two algorithmic solutions to tackle the cold start problem for new users and new items.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2022)
Article
Computer Science, Theory & Methods
Gang Wang, Qin Wang, Shiping Chen
Summary: This article comprehensively reviews the current progress of blockchain interoperability, explores the general principles and procedures of interoperable blockchain systems, compares state-of-the-art systems, and identifies critical challenges and potential research directions.
ACM COMPUTING SURVEYS
(2023)
Article
Engineering, Electrical & Electronic
Zeyu Wang, Xiongfei Li, Shuang Yu, Haoran Duan, Xiaoli Zhang, Jizheng Zhang, Shiping Chen
Summary: Multifocus image fusion (MFIF) is an efficient way to improve the visual effect of images with partial focus defects, and it is of great significance in the field of image enhancement. In this study, an edge-sensitive model for MFIF is presented, taking into account the correlation between salience object detection (SOD) and MFIF. Additionally, a randomized approach is proposed to generate massive training sets and pseudo-labels based on limited unlabeled data.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2023)
Article
Computer Science, Software Engineering
Jianmao Xiao, Zhipeng Xu, Donghua Zhang, Shiping Chen, Chenyu Liu, Zhiyong Feng, Guodong Fan, Chuying Ouyang
Summary: This paper proposes a research framework based on community mining to investigate the evolution process and influencing factors of mobile software ecosystems. By analyzing evolution events and crucial factors in different periods, the healthy operation of mobile software ecosystems can be maintained and improved.
JOURNAL OF SOFTWARE-EVOLUTION AND PROCESS
(2023)
Article
Computer Science, Hardware & Architecture
Shoulu Hou, Wei Ni, Kailan Zhao, Bo Cheng, Shuai Zhao, Zhiguo Wan, Xiulei Liu, Shiping Chen
Summary: This article presents a decentralized, three-timescale, online optimization approach that can significantly improve the energy efficiency of multicore micro data centers.
IEEE TRANSACTIONS ON SUSTAINABLE COMPUTING
(2023)
Proceedings Paper
Computer Science, Software Engineering
Yanli Li, Chongbin Ye, Huaming Chen, Shiping Chen, Minhui Xue, Jun Shen
Summary: In recent years, there has been rapid development in machine learning-based software service solutions, specifically for source code. However, the impact of source code engineering on these models is often overlooked. This study evaluates different parsing tools for their impact on a Code2Vec model's prediction task for method names in Java language. The results show that ASTs generated by different parsing tools vary significantly in terms of source code structures and contents, which can significantly affect the model's performance. Therefore, the selection of appropriate parsing tools during data pre-processing is crucial for machine learning models implemented in software services.
2023 IEEE INTERNATIONAL CONFERENCE ON SOFTWARE SERVICES ENGINEERING, SSE
(2023)
Proceedings Paper
Engineering, Electrical & Electronic
Weiliang Zhao, Anthony Heng, Luke Rosenberg, Si Tran Nguyen, Len Hamey, Mehmet Orgun
Summary: Inverse synthetic aperture radar (ISAR) is increasingly used in airborne maritime radar for noncooperative target imaging and classification. Traditional classification methods are limited by their reliance on geometric features extracted from images of known targets, which hampers their ability to classify unknown vessels. To address this challenge, this study proposes a transfer learning approach combined with an output layer called OpenMax. By comparing the new classification results with traditional methods and a three-layer Convolutional Neural Network (CNN) using a dataset of small vessels, it is observed that the use of OpenMax significantly improves classification performance for vessels from unknown classes.
2022 IEEE RADAR CONFERENCE (RADARCONF'22)
(2022)
Review
Chemistry, Physical
Antonella Sola, Yilin Sai, Adrian Trinchi, Clement Chu, Shirley Shen, Shiping Chen
Summary: Additive manufacturing (AM) is evolving towards industrial production, providing customized components for aerospace, defense, and biomedicine. Adding a tagging feature to AM parts is important for logistics, certification, and anti-counterfeiting. Materials engineers are researching the preferred tag types for different objects and how to modify existing materials and 3D printing hardware to create such tags.