November 2024
A comprehensive exploration and interpretability of Android malware with explainable deep learning techniques
2024, vol.16 , no.4, pp. 117-128
Article [2024-04-11]
This study introduces an innovative approach to tackle evolving Android malware threats using explainable artificial intelligence (XAI) methods combined with deep learning techniques. The framework enhances detection accuracy and provides interpretable insights into the model's decision-making process. The research utilizes the CICInvesAndMal2019 dataset for training with Deep Neural Network (DNN) techniques. It incorporates Shapley Additive Explanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) XAI techniques to refine the model's features and understand its predictions. This framework uses permissions and intents as static features from Android apps. The proposed framework reduces the execution time, reducing model loss to an impressively lower value of 0.26, and exhibits a commendable accuracy of 97.86%.
Diptimayee Sahu, Satya Narayan Tripathy. A comprehensive exploration and interpretability of Android malware with explainable deep learning techniques . International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 117-128.
Invisible boundaries: Balancing Image metadata privacy with forensic Imperatives
2024, vol.16 , no.4, pp. 105-116
Article [2024-04-10]
Metadata provides a wealth of information that is invaluable for digital forensic investigations, but it also poses significant privacy risks. The varying degrees of metadata retention across social media platforms have significant implications for digital forensic investigations. Metadata plays a crucial role in constructing timelines, verifying data authenticity, and providing evidence of interactions. In this research, we investigate the handling of EXIF metadata by various social media platforms and assess the implications for user privacy and digital forensic investigations. Our findings underscore the critical need for specialized software to manage and remove metadata from images uploaded to social media, preserving user privacy while retaining necessary data for legal and forensic purposes. To address these concerns, we implement a framework for encrypting and decrypting metadata, ensuring that sensitive information can be protected before images are uploaded to social media platforms. This research aims to encapsulate the findings and emphasize the need for balanced solutions that respect user privacy while supporting forensic investigations.
Abdullah Golam, Umar Albalawi. Invisible boundaries: Balancing Image metadata privacy with forensic Imperatives. International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 105-116.
Evaluation of supervised classification approach for DDoS threat detection in Software Defined Networks
2024, vol.16 , no.4, pp. 95-103
Article [2024-04-09]
Software defined networks (SDN) are the way of the future for networking, which separate the data and control planes of network devices to enable centralized control. SDN enhances network administration and security, allowing for programmability and improved performance. However, SDN is susceptible to attacks, particularly DDoS threats, which can overwhelm the network, restrict server access, and deplete network resources. This study offers a technique based on ML for recognizing DDoS threats in SDN. Various of ML models are assessed and trained using a publicly available SDN-specific dataset. These models’ performance is examined across different attack types, with results presented using parameters including F1_score, recall, precision and accuracy. The findings indicate the efficiency of ML techniques in identifying and mitigating against DDoS assaults in SDN environments.
Rakesh V. S., Vasanthakumar G. U.. Evaluation of supervised classification approach for DDoS threat detection in Software Defined Networks . International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 95-103.
A comprehensive framework for real-time malware detection and monitoring in production
2024, vol.16 , no.4, pp. 85-94
Article [2024-04-08]
This paper proposes a comprehensive framework for real-time malware detection and monitoring tailored to operational systems. Leveraging advanced machine learning algorithms, our framework integrates continuous monitoring mechanisms to ensure timely detection and response to emerging threats. The framework emphasizes regular assessment of model performance using metrics such as the Population Stability Index (PSI), ensuring models remain effective and adaptive to evolving malware patterns. By deploying models within the production environment, the framework enables regular evaluation and adaptation, enhancing the robustness and reliability of the detection system. Our results demonstrate the framework’s efficacy in providing a scalable and efficient solution for real-time malware detection and monitoring, contributing to improved cybersecurity posture in dynamic and high-risk environments.
Elshan Baghirov. A comprehensive framework for real-time malware detection and monitoring in production . International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 85-94.
CNN based multi-label image classification for presentation recommender system
2024, vol.16 , no.4, pp. 73-84
Article [2024-04-07]
Recommender systems are super popular nowadays for giving personalized recommendations to users. The paper focuses on a special type of recommender system aimed to advise an improvement of a user’s presentation based on slides classification. Classification is a fundamental task in information extraction and retrieval with a goal to assign given resource to a predefined class. In some cases, as in slides’ categorization, resources can belong to multiple classes, leading to a multi-labelled classification problem. Two different approaches for solving a multi-labelled image classification problem for the purpose of a presentation recommender system are suggested and evaluated in the paper. They are based on problem transformation and algorithm adaptation strategies and utilize a convolutional neural network for model training.
Maria Vlahova-Takova, Milena Lazarova. CNN based multi-label image classification for presentation recommender system. International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 73-84.
Application of the compositional machine learning dynamic classification algorithm for natural language sentiment analysis
2024, vol.16 , no.4, pp. 61-72
Article [2024-04-06]
The use of artificial intelligence methods for extract meaning from natural text is a scientific and practical field that is increasingly important in the context of creating software tools to interact with humans. Social networks generate a huge amount of textual information, the classification and proper processing of which can serve to counter crimes such as hate speech or provide useful information about events and trends in society. This paper presents an application of the author's Compositional Machine Learning Dynamic Classification (CoMLDC) algorithm. The algorithm dynamically classifies the emotional attitudes from Google’s dataset GoEmotions collected from communications of users in a social network. The results of the research show the advantages of the dynamic multi-class classification of the presented algorithm, over the classic algorithms for machine learning classification. The main advantage over the existing algorithms for multi-class, multi-label classification is finding a dynamic set of classes to which a given text message to be assigned without having to train a separate classifier for each class or set a constant number of classes. Instead, the classification result produced by the algorithm is achieved based on mathematical transformations and calculations.
Veneta Yosifova. Application of the compositional machine learning dynamic classification algorithm for natural language sentiment analysis. International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 61-72.
An online engineering education framework based on the predictors of adaptability and fuzzy inference system
2024, vol.16 , no.4, pp. 49-60
Article [2024-04-05]
Online engineering education utilizes the internet and information communication technologies as media of learning. Unfortunately, there is a dearth of frameworks developed that address the issues of online learning based on predictors of adaptability of engineering students and faculty members. This paper aims to identify the issues of online learning, establish the causal-effect relationship between socio-demographic factors and adaptability, and propose an Artificial Intelligence-driven framework using mixed methods. The author engaged 886 engineering students and 61 faculty members from the Technological University of the Philippines in Manila and Taguig Campuses in the Philippines. Based on test results, fewer disruptions, more ICT devices, lower monthly internet bills, faster internet speed, and age significantly predict the students’ adaptability to online learning. Meanwhile, faculty members with faster internet speed and higher monthly income have significantly higher adaptability to online learning. The author proposed a framework using a Fuzzy Inference System, which can be used for an accurate and timely decision-making process. Further test results confirmed that the framework is consistently accurate and significantly faster than the conventional method Hence, the proposed framework is a viable decision-making tool for large datasets and complex use cases in online engineering education.
Ralph Sherwin A. Corpuz. An online engineering education framework based on the predictors of adaptability and fuzzy inference system. International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 49-60.
On variants of state-dependent queueing systems for egress modelling
2024, vol.16 , no.4, pp. 37-48
Article [2024-04-04]
An M/G/c/c state dependent network is a quantitative model for replicating and analyzing the behavior of occupants travelling through a network. The model, however, assumes that its inter-arrival times are independent and exponentially distributed, which inadequately reflects the dynamic and high-stakes nature in emergency situations. This paper simulates the M/G/c/c model for emergency evacuations using Erlang-k distributions—representing the real-world arrival patterns of evacuees in a more controlled and less random arrivals—and correlated inter-arrival times—common during evacuations as occupants moving in clusters due to panic or structural flow patterns. To achieve this, an M/G/c/c simulation model was first developed using Arena simulation software. The model was used to analyze the impact of various arrival rates on system performance metrics. Consequently, the performance metrics were compared with those obtained by replacing exponential inter-arrival times with Erlang-k distributions and independent arrivals with correlated cluster arrivals. The results show that Erlang-k distributions lead to better performance and smoother flow since the arrival is more controlled, while correlated arrivals increase congestion.
Ruzelan Khalid , Gerrit K. Janssens . On variants of state-dependent queueing systems for egress modelling. International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 37-48.
Distilled and fused style embedding for authorship attribution with weighted multi-kernel ensemble model
2024, vol.16 , no.4, pp. 25-36
Article [2024-04-03]
In this age of Artificial Intelligence, the dispute over ownership of the web contents are increasing. In forensic studies and plagiarism cases, determining the author of a document undoubtedly plays a crucial role. The task of Authorship Attribution (AA) involves learning the style of the writers based on available documents to predict the ownership of unknown documents. Hence, capturing the style of writing is primary besides challenging. In this work, key features are systematically extracted, multiple deep-learning models are trained on various aspects of the text, and two novelties: Distilled and Fused Style Embedding (DFSE) and a multi-kernel ensemble model with kernels weighted based on Z-scores are proposed. The performance of the proposed model is superior to the baseline model and many contemporary models for in-genre and multi-topic datasets.
B. Lavanya, R. Sowmiya. Distilled and fused style embedding for authorship attribution with weighted multi-kernel ensemble model . International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 25-36.
Tools for designing complex software systems based on special linguistic constructions and algorithms for their processing
2024, vol.16 , no.4, pp. 15-24
Article [2024-04-02]
The article presents an approach to the design of complex software systems based on special linguistic constructions in the form of Backus-Nour forms and syntactic diagrams. The goal is to create special mathematical and software that would formally describe the components of the designed software system for the study of a complex object, as well as visually form the structure and algorithmic component of the system. As a result, a formal language for describing discrete-continuous objects has been developed, which ensures the unification of the display of static and dynamic objects by introducing a generalized "element" construction; special mathematical software has been created for the formal description of complex systems, allowing to obtain syntactic constructions of minimal complexity due to the orthogonality of linguistic means; algorithms for processing language constructions have been created that take into account the parallelism of the processes of the studied objects and ensure the automated creation of a class system of special software.
O. Ja. Kravets, I. A. Aksenov, Yu. V. Redkin, P. A. Rahman, D. I. Mutin, Amoa Kouadio-kan Armel Geoffroy, M. A. Ermolova. Tools for designing complex software systems based on special linguistic constructions and algorithms for their processing. International Journal on Information Technologies and Security, vol.16 , no.4, 2024, pp. 15-24.