Skripsi
MULTI-CLASSIFICATION SERANGAN SIBER MENGGUNAKAN METODE SUPPORT VECTOR MACHINE (SVM).
Cyberattacks are unauthorized attempts to track and disrupt the operation of communication systems and control systems by exploiting the security weaknesses of communication networks. Efforts to protect these systems can be made using multi-classification. Multi-classification is a classification that can understand the relationship between cyber-attacks so as to increase the ability of recognition and overcome the limitations related to classification. The method used in this research is Support Vector Machine (SVM) using the One Against one-SVM(OAO-SVM) and One against All-SVM(OAA-SVM) approaches so that the research results can be compared regarding accuracy and computing time. This research uses four types of datasets namely CSE-CIC-IDS2018, ISCXIDS2012, NSL-KDD, and KDD CUP 1999. By using validation on training and testing data from 20% to 80%. So that the best results were obtained on the CSE-CIC-IDS2018, ISCXIDS2012, and KDD CUP 1999 datasets using the SVM method with accuracy of 99.18%, 99.82%, and 99.92%. Then using the OAO-SVM method on the NSL-KDD dataset with the accuracy obtained is 99.66% with more efficient computing time on each dataset used. While the OAA-SVM method has lower accuracy with longer computation time.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2407002787 | T144076 | T1440762024 | Central Library (REFERENCES) | Available but not for loan - Not for Loan |
No other version available