Naive Bayes modification for intrusion detection system classification with zero probability

Kurniawan, Yogiek Indra; Razi, Fakhrur; Nofiyati, Nofiyati; Wijayanto, Bangun; Hidayat, Muhammad Luthfi

doi:10.11591/eei.v10i5.2833

Cited by 12 publications

(4 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The authors in [22], focused on enhancing the accuracy of GNB. The dataset selected for the research was KDD 99, which was cleaned before deciding the essential features using correlation-based feature selection (CSF).…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Iterative Feature Selection-Based DDoS attack Prevention Approach in Cloud

Naiem

Khedr

Idrees

et al. 2023

Int. j. electr. comput. eng. syst. (Online)

View full text Add to dashboard Cite

Distributed Denial of Service (DDOS) attacks aim to exploit the capacity and performance of a network's infrastructure, making the cloud environment one of the biggest targets for attackers. Many efforts are being made in the field of technology to prevent them from disrupting the services provided. Machine Learning techniques are a means to protect against DDOS attacks. Data preprocessing, feature selection, and classifiers are the main components of any prevention framework. The focus of this study is to find and enhance the feature selection approach for increasing the accuracy of the classifiers in detecting DDOS attacks from regular traffic. We used four different techniques, including Pearson Correlation Coefficient (PCC), Random Forest Feature Importance (RFFI), Mutual information (MI), and Chi-squared(X2) measure which we tested on different classifiers. The first selection approach was based on the feature’s independency level then the second iteration was based on the feature’s importance. We also examined the claim of dropping attacks from the dataset for better accuracy. The best performing set of features was from using PCC and RFFI together for feature selection with average accuracy and precision of 99.27% and 97.60%, which is higher than the use of PCC for both measures by almost 2%. The accuracy is also higher by nearly 12% from the same approach dropping 50% of the attacks.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

“…The NB is based on the hypothesis that features are independent. This classifier is considered one of the simple and easily implementable techniques for supervised machine learning classification [22] [20].…”

Section: Ddos Attacks Detection and Data Classificationmentioning

confidence: 99%

Iterative Feature Selection-Based DDoS attack Prevention Approach in Cloud

Naiem

Khedr

Idrees

et al. 2023

Int. j. electr. comput. eng. syst. (Online)

View full text Add to dashboard Cite

show abstract

“…Usually used for prediction, detection, and classification or recommendation purposes. Several supervised learning algorithms were used for representative classification methods including decision tree, naive Bayes [19], [20] and SVM [21] as well as ensemble method (RF) [22]. In this study, we used RF to classify SARS-Cov-2 DNA sequences by country.…”

Section: Classification Algorithmmentioning

confidence: 99%

Profiling DNA Sequence of SARS-Cov-2 Virus Using Machine Learning Algorithm

2022

View full text Add to dashboard Cite

Corona virus disease-19 (COVID-19) is growing rapidly because it is an infectious disease. This disease is caused by a virus belonging to the type of DNA virus with very diverse genetics. This study proposes a feature extraction method using k-mer to obtain nucleotide frequencies in protein coding. In profiling viral DNA sequences, this study proposes to obtain similarity by country using hierarchical k-means, where the results are averaged by the hierarchical clustering method and then find the initial cluster center. The experimental results show that the silhouette, purity, and entropy are 0.867, 0.208, and 0.892, respectively. Then, we apply the Gini index feature selection to find the important components as characteristics in each country. The selected components are implemented using the ensemble method, Random Forest, to evaluate their performance. The experimental results showed high performance, including sensitivity, accuracy, specificity, and area under the curve (AUC).

show abstract

“…Gambar 2. Konsep Kerja SIDS [1] (Sumber: Khraisat et al, 2019) Peneliti sebelumnya terkait alert IDS dengan klasifikasi Naïve Bayes antara lain dilakukan dengan: KDD99 dataset dan teknik PCA [7], AIDS dan teknik Correlation-Based Feature Selection [8], NSL-KDD dataset [9], zero probability [10], NSL-KDD dataset dan teknik PCA ditambah SVM [11], KNN sebagai perbandingan [12], diskritisasi variabel [13], Gaussian Naïve Bayes dan Kyoto dataset [14], HND dan KDD99 dataset [15], MNBIDS dan KDD99 dataset [16].…”

Section: Pendahuluanunclassified

Naïve Bayes dan Confusion Matrix untuk Efisiensi Analisa Intrusion Detection System Alert

Suryadewiansyah

Tju

2022

TEKNOSI

View full text Add to dashboard Cite

Banyaknya malware menyebabkan IDS (Intrusion Detection System) dituntut menyesuaikan diri semakin kompleks sehingga mahal dan membebani perusahaan yang menggunakannya. Sistem yang berbasis teknologi Host-based IDS dan Signatured-based IDS sudah banyak digunakan namun hanya mampu mendeteksi serangan yang sudah diketahui sebelumnya, untuk memperbaiki kinerjanya perlu dilakukan analisa pada data log berdasarkan alert yang diberikan. Teknik klasifikasi Naïve Bayes digunakan untuk membantu meningkatkan efisisensi dan efektifitas analisa tersebut. Penelitian ini dilakukan dengan mengambil empat langkah bagian dari metodologi SKKNI (Standar Kompetensi Kerja Nasional Indonesia) No.299 tahun 2020, Artificial Intelligence, sub bidang Data Science, yaitu data understanding, data preparation, modeling, dan model evaluation. Dataset dari penyedia layanan IDS sebanyak 575 data yang dibagi menjadi 515 data latih dan 60 data uji. Hasil evaluasi data uji dengan confusion matrix diperoleh pengukuran metrik accuracy 0,87, recall 0,89, precision 0,83, dan F-Measure 0,86. Adanya FP (False Positive) dan FN (False Negatif), keduanya sangat penting bagi penguna IDS untuk meningkatkan kualitas layanan kepada pelanggan dan mengurangi resiko akibat adanya intrusi. FP dan FN menjadi fokus dalam melakukan analisa log alert dari IDS sehingga tidak perlu menganalisa keseluruhan data, berdampak memberikan hasil 85% lebih efektif dan berkontribusi pada efisiensi tenaga dan waktu bagi tim keamanan suatu peruasahaan pengguna IDS. Selain itu didapat bahwa sekitar 50% data IDS adalah intrusi atau pengganggu lainnya.

show abstract

Naive Bayes modification for intrusion detection system classification with zero probability

Cited by 12 publications

References 25 publications

Iterative Feature Selection-Based DDoS attack Prevention Approach in Cloud

Iterative Feature Selection-Based DDoS attack Prevention Approach in Cloud

Profiling DNA Sequence of SARS-Cov-2 Virus Using Machine Learning Algorithm

Naïve Bayes dan Confusion Matrix untuk Efisiensi Analisa Intrusion Detection System Alert

Contact Info

Product

Resources

About