• International journal of advanced smart convergence
• /
• 제12권1호
• /
• pp.64-69
• /
• 2023

We devise a collaboration construction method based on the SPRT (Sequential Probability Ratio Test) for malware detection in IoT. In our method, high-end IoT nodes having capable of detecting malware and generating malware signatures harness the SPRT to give a reward of malware signatures to low-end IoT nodes providing useful data for malware detection in IoT. We evaluate our proposed method through simulation. Our simulation results indicate that the number of malware signatures provided for collaboration is varied in accordance with the threshold for fraction of useful data.

• International journal of advanced smart convergence
• /
• 제10권3호
• /
• pp.163-171
• /
• 2021

Recently, Machine Learning-based visualization approaches have been proposed to combat the problem of malware detection. Unfortunately, these techniques are exposed to Adversarial examples. Adversarial examples are noises which can deceive the deep learning based malware detection network such that the malware becomes unrecognizable. To address the shortcomings of these approaches, we present Block-matching and 3D filtering (BM3D) algorithm and deep image prior based denoising technique to defend against adversarial examples on visualization-based malware detection systems. The BM3D based denoising method eliminates most of the adversarial noise. After that the deep image prior based denoising removes the remaining subtle noise. Experimental results on the MS BIG malware dataset and benign samples show that the proposed denoising based defense recovers the performance of the adversarial attacked CNN model for malware detection to some extent.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권4호
• /
• pp.1799-1818
• /
• 2018

Malware detections continue to be a challenging task as attackers may be aware of the rules used in malware detection mechanisms and constantly generate new breeds of malware to evade the current malware detection mechanisms. Consequently, novel and innovated malware detection techniques need to be investigated to deal with this circumstance. In this paper, we propose a new secure malware detection system in which API call fragments are used to recognize potential malware instances, and these API call fragments together with the homomorphic encryption technique are used to construct a privacy-preserving Naive Bayes classifier (PP-NBC). Experimental results demonstrate that the proposed PP-NBC can successfully classify instances of malware with a hit-rate as high as 94.93%.

• International Journal of Computer Science & Network Security
• /
• 제21권6호
• /
• pp.286-293
• /
• 2021

Recently, the quick development rate of apps in the Android platform has led to an accelerated increment in creating malware applications by cyber attackers. Numerous Android malware detection tools have utilized conventional signature-based approaches to detect malware apps. However, these conventional strategies can't identify the latest apps on whether applications are malware or not. Many new malware apps are periodically discovered but not all malware Apps can be accurately detected. Hence, there is a need to propose intelligent approaches that are able to detect the newly developed Android malware applications. In this study, Radial Basis Function (RBF) networks are trained using known Android applications and then used to detect the latest and new Android malware applications. Initially, the optimal permission features of Android apps are selected using Information Gain Ratio (IGR). Appropriately, the features selected by IGR are utilized to train the RBF networks in order to detect effectively the new Android malware apps. The empirical results showed that RBF achieved the best detection accuracy (97.20%) among other common machine learning techniques. Furthermore, RBF accomplished the best detection results in most of the other measures.

• International journal of advanced smart convergence
• /
• 제13권3호
• /
• pp.41-47
• /
• 2024

In this paper, we deal with a game theoretic problem to explore interactions between evasive Artificial Intelligence (AI) malware and detectors in Internet of Things (IoT). Evasive AI malware is defined as malware having capability of eluding detection by exploiting artificial intelligence such as machine learning and deep leaning. Detectors are defined as IoT devices participating in detection of evasive AI malware in IoT. They can be separated into two groups such that one group of detectors can be armed with detection capability powered by AI, the other group cannot be armed with it. Evasive AI malware can take three strategies of Non-attack, Non-AI attack, AI attack. To cope with these strategies of evasive AI malware, detector can adopt three strategies of Non-defense, Non-AI defense, AI defense. We formulate a Bayesian game theoretic model with these strategies employed by evasive AI malware and detector. We derive pure strategy Bayesian Nash Equilibria in a single stage game from the formulated Bayesian game theoretic model. Our devised work is useful in the sense that it can be used as a basic game theoretic model for developing AI malware detection schemes.

• 정보보호학회논문지
• /
• 제29권4호
• /
• pp.775-784
• /
• 2019

신규 및 변종 악성코드의 발생으로 모바일, IoT, windows, mac 등 여러 환경에서 악성코드 침해 공격이 지속적으로 증가하고 있으며, 시그니처 기반 탐지의 대응만으로는 악성코드 탐지에 한계가 존재한다. 또한, 난독화, 패킹, Anti-VM 기법의 적용으로 분석 성능이 저하되고 있는 실정이다. 이에 유사성 해시 기반의 패턴 탐지 기술과 패킹에 따른 파일 분류 후의 정적 분석 적용으로 기계학습 기반 악성코드 식별이 가능한 시스템을 제안한다. 이는 기존에 알려진 악성코드의 식별에 강한 패턴 기반 탐지와 신규 및 변종 악성코드 탐지에 유리한 기계학습 기반 식별 기술을 모두 활용하여 보다 효율적인 탐지가 가능하다. 본 연구 결과물은 정보보호 R&D 데이터 챌린지 2018 대회의 AI기반 악성코드 탐지 트랙에서 제공하는 정상파일과 악성코드를 대상으로 95.79% 이상의 탐지정확도를 도출하여 분석 성능을 확인하였다. 향후 지속적인 연구를 통해 패킹된 파일의 특성에 맞는 feature vector와 탐지기법을 추가 적용하여 탐지 성능을 높이는 시스템 구축이 가능할 것으로 기대한다.

• Journal of Information Processing Systems
• /
• 제12권4호
• /
• pp.644-660
• /
• 2016

The real-time detection of malware remains an open issue, since most of the existing approaches for malware categorization focus on improving the accuracy rather than the detection time. Therefore, finding a proper balance between these two characteristics is very important, especially for such sensitive systems. In this paper, we present a fast portable executable (PE) malware detection system, which is based on the analysis of the set of Application Programming Interfaces (APIs) called by a program and some technical PE features (TPFs). We used an efficient feature selection method, which first selects the most relevant APIs and TPFs using the chi-square ($KHI^2$) measure, and then the Phi (${\varphi}$) coefficient was used to classify the features in different subsets, based on their relevance. We evaluated our method using different classifiers trained on different combinations of feature subsets. We obtained very satisfying results with more than 98% accuracy. Our system is adequate for real-time detection since it is able to categorize a file (Malware or Benign) in 0.09 seconds.

• 한국컴퓨터정보학회논문지
• /
• 제20권6호
• /
• pp.37-42
• /
• 2015

본 논문에서는 악성코드에 대한 피해를 실시간으로 탐지하고 차단하기 위해 모바일 내부에 악성링크에 대한 데이터베이스를 저장하고 또한 악성링크 탐지 엔진을 통해 웹 서비스를 통제함으로 인해 보다 안전한 모바일 환경을 제공하고자 한다. 최근 모바일 환경에서의 악성코드는 PC 환경 못지않게 기승을 부리고 있으며 새로운 위협이 되고 있다. 특히 모바일 특성상 악성코드의 피해는 사용자의 금전적인 피해로 이어진다는 것이 더 중요한 이유이다. 이러한 사이버 범죄를 어떻게 예방하고 실시간으로 차단할 수 있을 것 인지에 대해 많은 연구가 진행되고 있지만 초보적인 수준에 불과한 실정이다. 추가적으로 SMS나 MMS를 통해 전달되는 스미싱도 탐지 및 차단할 수 있는 방안을 제안하고자 한다. 향후 모바일 사업자는 본 연구를 바탕으로 한 근본적인 대책을 수립하여 안전한 모바일 환경을 구축해야 할 것이다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권9호
• /
• pp.3258-3273
• /
• 2021

Malware is a severe threat to the computing system and there's a long history of the battle between malware detection and anti-detection. Most traditional detection methods are based on static analysis with signature matching and dynamic analysis methods that are focused on sensitive behaviors. However, the usual detections have only limited effect when meeting the development of malware, so that the manual update for feature sets is essential. Besides, most of these methods match target samples with the usual feature database, which ignored the characteristics of the sample itself. In this paper, we propose a new malware detection method that could combine the features of a single sample and the general features of malware. Firstly, a structure of Directed Cyclic Graph (DCG) is adopted to extract features from samples. Then the sensitivity of each API call is computed with Markov Chain. Afterward, the graph is merged with the chain to get the final features. Finally, the detectors based on machine learning or deep learning are devised for identification. To evaluate the effect and robustness of our approach, several experiments were adopted. The results showed that the proposed method had a good performance in most tests, and the approach also had stability with the development and growth of malware.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.2180-2197
• /
• 2019

With the proliferation of the Android malicious applications, malware becomes more capable of hiding or confusing its malicious intent through the use of code obfuscation, which has significantly weaken the effectiveness of the conventional defense mechanisms. Therefore, in order to effectively detect unknown malicious applications on the Android platform, we propose DroidVecDeep, an Android malware detection method using deep learning technique. First, we extract various features and rank them using Mean Decrease Impurity. Second, we transform the features into compact vectors based on word2vec. Finally, we train the classifier based on deep learning model. A comprehensive experimental study on a real sample collection was performed to compare various malware detection approaches. Experimental results demonstrate that the proposed method outperforms other Android malware detection techniques.