Garuda - Garba Rujukan Digital

Enhancing XGBoost Performance in Malware Detection through Chi-Squared Feature Selection Rosyada, Salma; Rafrastara, Fauzi Adi; Ramadhani, Arsabilla; Ghozi, Wildanil; Yassin, Warusia
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 13 No. 3 (2024): NOVEMBER
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v13i3.2293

The increasing prevalence of malware poses significant risks, including data loss and unauthorized access. These threats manifest in various forms, such as viruses, Trojans, worms, and ransomware. Each continually evolves to exploit system vulnerabilities. Ransomware has seen a particularly rapid increase, as evidenced by the devastating WannaCry attack of 2017 which crippled critical infrastructure and caused immense economic damage. Due to their heavy reliance on signature-based techniques, traditional anti-malware solutions struggle to keep pace with malware's evolving nature. However, these techniques face limitations, as even slight code modifications can allow malware to evade detection. Consequently, this highlights weaknesses in current cybersecurity defenses and underscores the need for more sophisticated detection methods. To address these challenges, this study proposes an enhanced malware detection approach utilizing Extreme Gradient Boosting (XGBoost) in conjunction with Chi-Squared Feature Selection. The research applied XGBoost to a malware dataset and implemented preprocessing steps such as class balancing and feature scaling. Furthermore, the incorporation of Chi-Squared Feature Selection improved the model's accuracy from 99.1% to 99.2% and reduced testing time by 89.28%, demonstrating its efficacy and efficiency. These results confirm that prioritizing relevant features enhances both the accuracy and computational speed of the model. Ultimately, combining feature selection with machine learning techniques proves effective in addressing modern malware detection challenges, not only enhancing accuracy but also expediting processing times.

OPTIMIZING ANDROID MALWARE DETECTION USING NEURAL NETWORKS AND FEATURE SELECTION METHOD Bintoro, Jevan; Rafrastara, Fauzi Adi; Latifah, Ines Aulia; Ghozi, Wildani; Yassin, Warusia
Jurnal Teknik Informatika (Jutif) Vol. 5 No. 6 (2024): JUTIF Volume 5, Number 6, Desember 2024
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2024.5.6.3898

Malware poses a serious threat to Android security systems. In recent years, Android malware has rapidly evolved, employing obfuscation techniques such as polymorphic and metamorphic. Unfortunately, signature-based malware detection cannot identify modern variants of Android malware. This study aims to compare various feature selection methods and machine learning algorithms to identify the most effective and efficient combination for classifying Android malware. The dataset used in this research is the Drebin dataset. Four classification algorithms are used in this comparison: Naive Bayes, Logistic Regression, Neural Network, and Random Forest. The best-performing algorithm is then implemented in three different scenarios: without feature selection, with Information Gain, and with Chi-Squared (X²). In the latter two scenarios, the appropriate number of features was selected using the backward elimination method. Both feature selections achieved the same performance, but Information Gain required fewer features. The evaluation metrics used in this study include AUC, accuracy, F1-score, training time, and testing time. Measuring training and testing time benefits the model by making it more efficient, thus allowing for faster detection in real-world applications. The results show that the combination of the Information Gain feature selection method and the Neural Network algorithm achieves the highest performance, with an accuracy and F1-Score of 98.6%. Additionally, this combination achieves a training time of 81.135 seconds and a testing time of 1.095 seconds. Compared to the Neural Network algorithm without feature selection, this combination results in a 17.7597 % reduction in training time and a 57.9977 % reduction in testing time while maintaining the same performance values. This research contributes to improving the speed and accuracy of malware detection systems, enhancing mobile security.

CLASSIFICATION OF TODDLER NUTRITIONAL STATUS USING SUPPORT VECTOR MACHINE AND RANDOM FOREST TECHNIQUES WITH OPTIMAL FEATURE SELECTION Widyawati, Femmi; Suhito, Hanif Pandu; Yassin, Warusia; Agus Santoso, Heru
Jurnal Teknik Informatika (Jutif) Vol. 5 No. 6 (2024): JUTIF Volume 5, Number 6, Desember 2024
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2024.5.6.4162

Nutritional problems in toddlers, such as stunting, wasting, being underweight, and obesity, are major challenges in monitoring toddler health in Indonesia because they can hurt toddler growth and development. Therefore, handling nutritional problems comprehensively, including prevention efforts and appropriate dietary interventions, is very important. This study aims to develop a toddler nutritional status classification model based on machine learning algorithms, namely Support Vector Machine (SVM) and Random Forest, by utilizing a toddler dataset obtained from Health Institutions in Indonesia containing 9,735 data. The model was designed using the Recursive Feature Elimination (RFE) technique for selecting relevant features and the Synthetic Minority Over-sampling Technique (SMOTE) to handle class imbalance. The results showed that the Random Forest algorithm performed best with 95% accuracy, 77% precision, 87% recall, and 81% f1-score. This study contributes to developing a machine learning-based approach to support a more effective nutritional monitoring system and enable more appropriate dietary interventions to address toddler health problems in Indonesia.

Enhancing XGBoost Performance in Malware Detection through Chi-Squared Feature Selection Rosyada, Salma; Rafrastara, Fauzi Adi; Ramadhani, Arsabilla; Ghozi, Wildanil; Yassin, Warusia
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 13 No. 3 (2024): NOVEMBER
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v13i3.2293

The increasing prevalence of malware poses significant risks, including data loss and unauthorized access. These threats manifest in various forms, such as viruses, Trojans, worms, and ransomware. Each continually evolves to exploit system vulnerabilities. Ransomware has seen a particularly rapid increase, as evidenced by the devastating WannaCry attack of 2017 which crippled critical infrastructure and caused immense economic damage. Due to their heavy reliance on signature-based techniques, traditional anti-malware solutions struggle to keep pace with malware's evolving nature. However, these techniques face limitations, as even slight code modifications can allow malware to evade detection. Consequently, this highlights weaknesses in current cybersecurity defenses and underscores the need for more sophisticated detection methods. To address these challenges, this study proposes an enhanced malware detection approach utilizing Extreme Gradient Boosting (XGBoost) in conjunction with Chi-Squared Feature Selection. The research applied XGBoost to a malware dataset and implemented preprocessing steps such as class balancing and feature scaling. Furthermore, the incorporation of Chi-Squared Feature Selection improved the model's accuracy from 99.1% to 99.2% and reduced testing time by 89.28%, demonstrating its efficacy and efficiency. These results confirm that prioritizing relevant features enhances both the accuracy and computational speed of the model. Ultimately, combining feature selection with machine learning techniques proves effective in addressing modern malware detection challenges, not only enhancing accuracy but also expediting processing times.

Title

Found 4 Documents
Search

Abstract

Abstract

Abstract

Abstract

Title Search

Found 4 Documents Search

Abstract

Abstract

Abstract

Abstract

Title

Found 4 Documents
Search