Text
DETEKSI ANOMALI FILE PDF MALWARE PADA LAYANAN AGREGATOR GARBA RUJUKAN DIGITAL (GARUDA) DENGAN ALGORITMA DECISION TREE
Portable Document Format (PDF) is a document exchange media that is very vulnerable to malicious attacks, namely Malware PDF. One of the services that most often use PDF files as a medium is a scientific publication service Garba Rujukan Digital (GARUDA). Therefore, research was conducted using static analysis methods for each PDF and data extraction using PDFiD. Based on these research, it found an oddity or anomaly to some PDF files so that the dataset is divided into three classes, namely PDF benign, PDF anomaly, and PDF malware. The generated dataset in this research is a dataset with imbalanced conditions and used Synthetic Minority Oversampling Technique (SMOTE) and NearMiss to balance the data. To classify malware PDF file attacks used one of the well-known machine learning methods, Decision Tree Algorithm. Classification divided into two types, classification with the original dataset (imbalanced dataset conditions) and classification with balanced dataset conditions. Then to validate the accuracy of the classification model used cross validation method, Stratified K-Fold Cross Validation. Based on classification results, the best performance obtained by the average percentage of accuracy 99.83%, precision 99.83%, recall 99.83%, F1-score 99.84%, TNR (true negative rate) 99.92%, AUC (area under curve) 99.88%, and FPR (false positive rate) 0.001 and FNR (false negative rate) 0.002.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2207005454 | T85028 | T850282022 | Central Library (Referens) | Available but not for loan - Not for Loan |
No other version available