Browsing by Author "Islam, Takia"

Now showing 1 - 8 of 8

A Large-Scale Investigation to Identify the Pattern of App Component in Obfuscated Android Malwares
(Springer, 2020-06-15) Russel, Md. Omar Faruque Khan; Motiur Rahman, Sheikh Shah Mohammad; Islam, Takia
Number of smartphone users of android based devices is growing rapidly. Because of the popularity of the android market malware attackers are focusing in this area for their bad intentions. Therefore, android malware detection has become a demanding and rising area to research in information security. Researchers now can effortlessly detect the android malware whose patterns have formerly been recognized. At present, malware attackers commenced to use obfuscation techniques to make the malwares incomprehensible to malware detectors. For this motive, it is urgent to identify the pattern that is used by attackers to obfuscate the malwares. A large-scale investigation has been performed in this paper by developing python scripts to extract the pattern of app components from an obfuscated android malware dataset. Ultimately, the patterns in a matrix form has been established and stored in a Comma Separated Values (CSV) file which will conduct to the primary basis of detecting the obfuscated malwares.
A Large-Scale Investigation to Identify the Pattern of Permissions in Obfuscated Android Malwares
(Springer, 2020-07-30) Russel, Md. Omar Faruque Khan; Motiur Rahman, Sheikh Shah Mohammad; Islam, Takia
This paper represents a simulation-based investigation of permissions in obfuscated android malware. Android malware detection has become a challenging and emerging area to research in information security because of the rapid growth of android based smartphone users. To detect malwares in android, permissions to access the functionality of android devices play an important role. Researchers now can easily detect the android malwares whose patterns have already been identified. However, recently attackers started to use obfuscation techniques to make the malwares unintelligible. For that reason, it’s necessary to identify the pattern used by attackers to obfuscate the malwares. In this paper, a large-scale investigation has been performed by developing python scripts to extract the pattern of permissions from an obfuscated malwares dataset named Android PRAGuard Dataset. Finally, the patterns in a matrix form has been found and stored in a Comma Separated Values (CSV) file which will lead to the fundamental basis of detecting the obfuscated malwares.
An Investigation and Evaluation of N-Gram, TF-IDF and Ensemble Methods in Sentiment Classification
(Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST, 2020) Rahman, Sheikh Shah Mohammad Motiur; Biplob, Khalid Been Md. Badruzzaman; Rahman, Md. Habibur; Sarker, Kaushik; Islam, Takia
In the area of sentiment analysis and classification, the performance of the classification tasks can be varied based on the usage of text vectorization and feature extraction methods. This paper represents a detailed investigation and analysis of the impact on feature extraction methods to attain the highest classification accuracy of the sentiment from user reviews. Unigram, Bigram and Trigram are applied as n-gram vectorization models with TF-IDF features extraction method individually. Accuracy, misclassification rate, Receiver Operating Characteristics (ROC) and recall-precision are used in this study to evaluate which are counted as the most important performance measurement parameters in machine learning based approaches. Parameters are measured by the output obtained from Bagged Decision Tree (BDT), Random Forest (RF), Ada Boost (ADA), Gradient Boost (GB) and Extra Tree (ET). The outcomes of this study is to find out the best fitted combination of term frequency–inverse document frequency (TF-IDF) and n-grams for different data size.
Evaluation of N-gram Based Multi-layer Approach to Detect Malware in Android
(Procedia Computer Science, Elsevier, 2020) Islam, Takia; Rahman, Sheikh Shah Mohammad Motiur; Hasan, Md. Aumit; Rahaman, Abu Sayed Md. Mostafizur; Jabiullah, Md. Ismail
N-gram techniques usually used in Natural Language Processing (NLP). Those techniques along with stacked generalization has been experimented and assessed in the field of android malware detection. Beacuse of the rapidly growing of android users, android malware has become most popular among the attackers. Android malware has become gigantic topics in information security. Various security researchers have already started to propose intelligency based android malware detection. In this paper, a details investigation has been performed to evaluate the effectiveness of unigram, bigram and trigram with stacked generalization. It’s been found that with stacking, unigram provides more than 97% of accuracy which is highest detection rate against bigram and trigram. In level 1, Extra Tree (ET), Random Forest (RF) and Gradient Boosting (GB) are used. As a final predictor and meta estimator eXtreme Gradient Boosting (XGBoost) is used. A strong basement to use n-gram techniques in developing android malware detection has been determined from this study.
IntAnti-Phish
(Scopus, 2021) Rahman, Sheikh Shah Mohammad Motiur; Gope, Lakshman; Islam, Takia; Alazab, Mamoun
Among the cybercriminals, the popularity of phishing has been rapidly growing day by day. Therefore, phishing has become an alarming issue to solve in the field of cyber security. Many researchers have already proposed several anti-phishing approaches to detect phishing in terms of email, webpages, images, or links. This study also aimed to propose and implement an intelligent framework to detect phishing URLs (Uniform Resource Locator). It has been observed in this study that Back propagation Neural Network-based systems need to tune various hyper parameters to obtain the optimized output. With a maximum of two hidden layers along with 400 epochs can reach maximum accuracy of 0.93, the minimum mean squared error of 0.27, and also a minimum error rate of 0.07 which measurements lead this study to generate an optimized model for phishing detection. The detailed process of feature extraction and optimized model generation along with the detection of unknown URLs are considered and proposed during the development of IntAnti-Phish (An Intelligent Anti-Phishing Framework).
IntAnti-Phish
(Studies in Computational Intelligence , Springer, 2020-12-15) Rahman, Sheikh Shah Mohammad Motiur; Gope, Lakshman; Islam, Takia; Alazab, Mamoun
Among the cybercriminals, the popularity of phishing has been rapidly growing day by day. Therefore, phishing has become an alarming issue to solve in the field of cybersecurity. Many researchers have already proposed several anti-phishing approaches to detect phishing in terms of email, webpages, images, or links. This study also aimed to propose and implement an intelligent framework to detect phishing URLs (Uniform Resource Locator). It has been observed in this study that Backpropagation Neural Network-based systems need to tune various hyperparameters to obtain the optimized output. With a maximum of two hidden layers along with 400 epochs can reach maximum accuracy of 0.93, the minimum mean squared error of 0.27, and also a minimum error rate of 0.07 which measurements lead this study to generate an optimized model for phishing detection. The detailed process of feature extraction and optimized model generation along with the detection of unknown URLs are considered and proposed during the development of IntAnti-Phish (An Intelligent Anti-Phishing Framework).
Phish Stack
(Scopus, 2020) Rahman, Sheikh Shah Mohammad Motiur; Islam, Takia; Jabiullah, Md. Ismail
Stacked Generalization has been assessed and evaluated in the field of Phishing URLs detection. This field has become egregious area of information security. Recently, different phishing URLs detection systems have already proposed by several researchers. But due to the lack of proper machine learning algorithm selection, the performance of those systems can be affected. A details investigation on individual machine learning classifiers on level 1 and final prediction from level 2 along with three real datasets have been presented on this paper. The performance has been evaluated by precision-recall curve, AUC-ROC curve, accuracy, misclassification rate and mean absolute error (MAE). The best AUC area obtained from Random Forest and Multi Layer Perceptron (MLP) individually. But stacked generalization provides higher accuracy of 97.44% with numeric feature set in binary classification and in multiclass feature set (dataset three), provides the performance with 97.86% of accuracy. Stacked generalization provides minimum error rate and MAE of 2.142857% with multiclass feature set which leads to a strong basement of developing an anti-phishing tools.
Towards A Blockchain Based Platform to Share the Datasets
(Daffodil International University, 2018-12-23) kumar, D K Tonoy; Islam, Takia
Conventional cloud or central authority based storage has depended solely on extensive capacity suppliers, who act as trusted third parties to exchange and store information. This model represents various issues including information accessibility, high operational expense, and data security. The blockchain is able to maintain transaction without a central authority, the blockchain is an imaginative innovation which opened ways to new applications for tackling various issues in the distributed environment. The primary objectives in this thesis creating secure data sharing platform where the data provider will share their research related data that the verified researcher can use by downloading and contribute to further research. This thesis considers the future prospects of blockchain technology and cryptographic ability in data storage. In this thesis, we proposed an architecture that leverage blockchain technology and provide secure data storage. The data provider shares their data into consortium blockchain network that consist of plaintext and encrypted form the same plaintext hash value via data provider private key (digital signature) and after completion the validating process then the encrypted data will be broadcast into public blockchain network and finally the data requester could request for their desired data via joining the public blockchain network. We have been trying to analysis our proposed architecture performance via comparison among existing blockchain based data storage and traditional database system