Article
Computer Science, Information Systems
Manuel Torres, Rafael Alvarez, Miguel Cazorla
Summary: Cybercriminals constantly develop new techniques to evade security measures, resulting in rapid evolution of malware. Detecting malware across multiple systems is challenging due to unique characteristics of each computing environment. Traditional signature-based malware detection has been replaced by modern approaches, such as machine learning and behavior-based threat detection. Researchers use these techniques to extract features from various data sources and feed them to models for accurate prediction.
Article
Computer Science, Interdisciplinary Applications
C. Catalano, A. Chezzi, M. Angelelli, F. Tommasi
Summary: The study critically analyzes the strengths and weaknesses of using CNN for static malware detection, starting from the conversion of binary executable files to pixel images. It aims to achieve fast and accurate malware classification by relying solely on the binary content of the file.
COMPUTERS IN INDUSTRY
(2022)
Article
Multidisciplinary Sciences
Daniel P. Furtado, Edson A. Vieira, Wildna Fernandes Nascimento, Kelly Y. Inagaki, Jessica Bleuel, Marco Antonio Zanata Alves, Guilherme O. Longo, Luiz S. Oliveira
Summary: Corals play a significant role in marine environments by providing habitat for various organisms, but climate change is causing them to lose color. Monitoring coral reefs is crucial for understanding their response to human impacts, and the Marine Ecology Laboratory at the Federal University of Rio Grande do Norte has developed a project that allows people to contribute to coral monitoring by sharing photos on social media. By using machine learning algorithms, they have improved the efficiency and accuracy of image analysis for coral monitoring.
Article
Computer Science, Hardware & Architecture
Pascal Maniriho, Abdun Naser Mahmood, Mohammad Jabed Morshed Chowdhury
Summary: This paper presents API-MalDetect, a new deep learning-based framework for automated detection of malware attacks in Windows systems. Experimental results show that API-MalDetect outperforms existing state-of-the-art techniques in terms of accuracy, precision, recall, F1-score, and AUC-ROC. The framework is effective in automatically identifying unique patterns from raw and long sequences of API calls to distinguish malware attacks from benign activities in Windows systems.
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Sanjeev Kumar, Kajal Panda
Summary: This paper proposes a novel malware detection and classification architecture based on image visualization using fine-tuned convolutional neural networks. The methodology involves using a pre-trained VGG16 model as a feature extractor and different feature selection methods to construct a feature map. The MLP classifier achieves the best accuracy in detecting malware.
APPLIED SOFT COMPUTING
(2023)
Article
Biology
Muhammad Najam Dar, Muhammad Usman Akram, Rajamanickam Yuvaraj, Sajid Gul Khawaja, M. Murugappan
Summary: EEG-based emotion classification enables reliable and meaningful human-computer interaction, with applications in entertainment consumption, brain-computer interface, and psychological healthcare. In this study, a new architecture combining a 1D-CRNN and ELM is proposed for robust emotion detection in Parkinson's disease patients. The method achieves high accuracy across different datasets and emotions.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Agronomy
Mohd Firdaus Ibrahim, Siti Khairunniza-Bejo, Marsyita Hanafi, Mahirah Jahari, Fathinul Syahir Ahmad Saad, Mohammad Aufa Mhd Bookeri
Summary: Rice is a primary food source for almost half of the global population, and Asia accounts for around 90% of rice production worldwide. The study developed a machine vision system using high-density images of planthoppers collected in the field to classify the species. Five deep CNN models were tested, and the ResNet-50 model achieved the best performance with high accuracy, precision, recall, and F1-score.
Article
Automation & Control Systems
Kamran Shaukat, Suhuai Luo, Vijay Varadharajan
Summary: This paper proposes a novel deep learning-based approach for malware detection, which combines the advantages of static and dynamic analysis to achieve better performance than conventional methods. It converts portable executable (PE) files into colored images and extracts deep features using a fine-tuned deep learning model. Malware is then detected based on these deep features using support vector machines (SVM). The proposed method eliminates the need for intensive feature engineering tasks and domain knowledge, and it is scalable, cost-effective, and efficient.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Theory & Methods
Junyang Qiu, Jun Zhang, Wei Luo, Lei Pan, Surya Nepal, Yang Xiang
Summary: Deep Learning (DL) is a disruptive technology that has revolutionized cyber security research, especially in the detection and classification of Android malware. While offering many advantages, DL faces challenges such as choice of architecture, feature extraction, and obtaining high-quality data.
ACM COMPUTING SURVEYS
(2021)
Article
Business, Finance
Stephane Goutte, Hoang-Viet Le, Fei Liu, Hans-Jorg von Mettenheim
Summary: This study explores the potential use of technical analysis as inputs for machine learning models, especially state-of-the-art deep learning algorithms, to generate trading signals. Through empirical research on five years of Bitcoin hourly data from 2017 to 2022, we confirm the potential of trading strategies using machine learning approaches and find that deep learning models, specifically recurrent neural networks, tend to outperform other models in time-series prediction.
FINANCE RESEARCH LETTERS
(2023)
Article
Engineering, Chemical
Muhammad Ashfaq Khan
Summary: Network attacks are a crucial problem in modern society, and developing effective intrusion detection systems is essential to mitigate the impact of malicious threats. Utilizing deep learning and machine learning techniques, researchers have designed a hybrid convolutional recurrent neural network intrusion detection system that achieves high accuracy in detecting malicious cyberattacks.
Article
Computer Science, Theory & Methods
Yi-Hsien Chen, Si-Chen Lin, Szu-Chun Huang, Chin-Laung Lei, Chun-Ying Huang
Summary: Malicious binaries have caused harm to individuals in terms of data and financial loss, and they are rapidly evolving. Security analysts and researchers face the task of identifying and reporting malicious behaviors within these binaries. This study aims to accelerate and automate the analysis process by identifying essential functions in malicious binaries.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2023)
Article
Computer Science, Hardware & Architecture
Sanjeev Kumar, B. Janet
Summary: This paper introduces a novel malware threat intelligence system (MTIS) to detect modern and real-world malware samples with better classification accuracy without the need for code reversing and domain expertise. By combining grayscale images and texture feature extraction methods, the proposed architecture is resilient to packed and encrypted malware samples.
Article
Energy & Fuels
Yiheng Pang, Liang Hao, Yun Wang
Summary: This study presents a machine learning approach using convolutional neural networks to analyze neutron radiography images and quantify liquid water content in polymer electrolyte membrane fuel cells. The results show that using low relative humidity inlet flow can significantly reduce water content, while counter-flow configuration increases water content compared to co-flow configuration.
Article
Computer Science, Information Systems
Olorunjube James Falana, Adesina Simon Sodiya, Saidat Adebukola Onashoga, Biodun Surajudeen Badmus
Summary: Recent outbreaks of pandemics have led to an increase in cyberattacks caused by malware. This study proposes a novel ensemble technique, called Mal-Detect, which combines Deep Convolutional Neural Network and Deep Generative Adversarial Neural Network to analyze, detect, and categorize malware. Experimental results demonstrate that Mal-Detect outperforms other state-of-the-art techniques with an accuracy of 99.8% in detecting malware.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Gianni D'Angelo, Francesco Palmieri
Summary: With the emergence of COVID-19, mobile health applications have become increasingly crucial. This study aims to enhance the performance of COVID-19 tracking apps by providing a human activity classifier based on Convolutional Deep Neural Network. Experimental results showed that the HAR-Images are effective features for human activity recognition.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Hardware & Architecture
Raffaele Cerulli, Ciriaco D'Ambrosio, Antonio Iossa, Francesco Palmieri
Summary: This paper introduces the Maximum Network Lifetime Problem (MLP) and its variant Maximum Lifetime Problem with Time Slots (MLPTS) in wireless sensor networks. Three different approaches are proposed and compared through extensive computational experiments, showing that the Carousel Greedy algorithm represents the best trade-off between the proposed methods and can significantly improve network lifetime.
JOURNAL OF SUPERCOMPUTING
(2022)
Article
Computer Science, Information Systems
Gianni D'Angelo, Francesco Palmieri, Antonio Robustelli
Summary: The paper proposes a new feature model called permission maps (Perm-Maps), which effectively classifies different malware families by combining information about Android permissions and their corresponding severity levels. The use of Perm-Maps, along with a training process based on federated logic, improves classification accuracy compared to other classifiers and allows for dealing with unbalanced training datasets.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Dun Li, Dezhi Han, Tien-Hsiung Weng, Zibin Zheng, Hongzhi Li, Han Liu, Arcangelo Castiglione, Kuan-Ching Li
Summary: Federated learning is a decentralized deep learning technology that collaboratively updates models without sharing data, but it faces challenges such as privacy, communication costs, system heterogeneity, and unreliable model uploads in practice. Integrating blockchain technology into federated learning to create the Blockchain-based federated learning framework can improve security, performance, and application scope.
Article
Computer Science, Theory & Methods
Gianni D'Angelo, David Della-Morte, Donatella Pastore, Giulia Donadel, Alessandro De Stefano, Francesco Palmieri
Summary: Diabetes mellitus is a global health problem, and its most debilitating complication, diabetic foot, increases the risk of hospitalization, morbidity, and mortality. This study presents a Genetic Programming-based approach called X-GPC, which provides a global interpretation of the diabetic foot ulcer diagnosis through a mathematical model. It also offers a consultable 3D graph for medical staff to understand patients' situations and make decisions for their healing. Experimental results show that the proposal achieves 100% accuracy in diagnosing diabetic foot, outperforming other state-of-the-art techniques.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2023)
Article
Computer Science, Information Systems
Bruno Carpentieri, Francesco Palmieri
Summary: The majority of compressed digital data on modern high-speed networks is directly linked to human activities, raising concerns about privacy protection and safeguarding digital multimedia contents. This paper explores a unified approach to compression and privacy by considering various types of digital data (text, images, sound, and hyperspectral images).
Article
Computer Science, Theory & Methods
Gianni D'Angelo, Eslam Farsimadan, Massimo Ficco, Francesco Palmieri, Antonio Robustelli
Summary: The emergence of new and sophisticated malware targeting Android-based IoT devices poses security risks and the need for effective detection models and strategies. Federated Learning-based solutions, which use Machine Learning models without sharing user data, are being developed. However, these methods are affected by non-independent and identically distributed data. Privacy-preserving approaches using Markov chains and associative rules are proposed to handle malware classification in the IoT scenario. The approach achieves high accuracy and comparable runtime performance with centralized methods.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2023)
Article
Computer Science, Theory & Methods
Roberto Pietrantuono, Massimo Ficco, Francesco Palmieri
Summary: This paper proposes a hybrid method to assess the survivability of an IoT system under resource-exhaustion attacks and optimize the preventive maintenance trigger period. The method combines measurements and model-based analysis to estimate resource consumption and simulate system behavior during attacks.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2023)
Article
Computer Science, Theory & Methods
Xiaolong Xu, Zexuan Fan, Marcello Trovati, Francesco Palmieri
Summary: This study proposes a local differential privacy (LDP) solution for multi-layer networks in edge computing scenarios, aiming to overcome limitations in key-value data heavy hitter identification and related frequency and mean estimation tasks. The proposed method optimizes the utility/performance of edge nodes and reduces communication and storage costs, while introducing an improved user grouping strategy. Experimental results show that the proposed method achieves better performance in heavy hitter identification, frequency, and mean estimations compared to other mechanisms.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2023)
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)
Proceedings Paper
Computer Science, Information Systems
Dajana Conte, Eslam Farsimadan, Leila Moradi, Francesco Palmieri, Beatrice Paternoster
Summary: This study proposes a numerical technique based on a hybrid of block-pulse functions and Chelyshkov polynomials to solve fractional delay differential equations. The suggested method's accuracy and efficiency are demonstrated using numerical examples.
COMPUTATIONAL SCIENCE AND ITS APPLICATIONS, ICCSA 2022, PT I
(2022)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Bruno Carpentieri, Francesco Palmieri
Summary: This paper explores a unified approach to compression and privacy in response to the significant increase in network traffic and the growing need for privacy protection. It presents a secure protocol for interactive data compression and a new algorithm for scrambling the Region of Interest (ROI) of an image.
EXTENDED REALITY, XR SALENTO 2022, PT I
(2022)
Proceedings Paper
Computer Science, Information Systems
Gianni D'Angelo, Francesco Palmieri, Antonio Robustelli
Summary: The outbreak of the COVID-19 pandemic has led to a significant increase in the use of mobile devices by employees worldwide to access corporate systems, making them more vulnerable to malicious applications. In this paper, a novel approach called API-Streams is proposed to minimize damages at runtime. Through the use of CNN-LSTM Autoencoders, the proposed approach achieves an average accuracy of 98% in video classification tasks.
MOBILE INTERNET SECURITY, MOBISEC 2021
(2022)
Article
Computer Science, Theory & Methods
Meriem Guerar, Luca Verderame, Mauro Migliardi, Francesco Palmieri, Alessio Merlo
Summary: A recent study has shown that malicious bots generated a significant portion of website traffic in 2019, posing a serious threat to businesses. In order to combat these bots, introducing CAPTCHA tests has become a common defense mechanism. Therefore, understanding the effectiveness of different CAPTCHA schemes is crucial. This paper provides an overview of the current research in the field of CAPTCHA schemes and introduces a new classification. It also summarizes various attack methods and discusses the limitations of different CAPTCHA schemes.
ACM COMPUTING SURVEYS
(2022)