scholarly journals Review the performance of the Bernoulli Naïve Bayes Classifier in Intrusion Detection Systems using Recursive Feature Elimination with Cross-validated selection of the best number of features

2021 ◽  
Vol 190 ◽  
pp. 564-570
Author(s):  
Mechetin Artur
Keyword(s):  
Intrusion Detection ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Intrusion Detection Systems ◽  
Recursive Feature Elimination ◽  
Naive Bayes Classifier ◽  
Bayes Classifier ◽  
Naïve Bayes Classifier ◽  
Detection Systems ◽  
Selection Of

scholarly journals K-Means Clustering and Naive Bayes Classification for Intrusion Detection

2016 ◽  
Vol 4 (1) ◽  
pp. 13-25 ◽  
Author(s):  
Z. Muda ◽  
W. Yassin ◽  
M.N. Sulaiman ◽  
N.I. Udzir
Keyword(s):  
Intrusion Detection ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Intrusion Detection Systems ◽  
False Alarms ◽  
Bayes Classifier ◽  
Detection Systems ◽  
Naive Bayes Classification ◽  
Naïve Bayes Classification ◽  
Accuracy Rates

Intrusion detection systems (IDS) effectively complement other security mechanisms by detecting malicious activities on a computer or network, and their development is evolving at an extraordinary rate. The anomaly-based IDS, which uses learning algorithms, allows detection of unknown attacks. Unfortunately, the major challenge of this approach is to minimize false alarms while maximizing detection and accuracy rates. To overcome this problem, we propose a hybrid learning approach through the combination of K-Means clustering and Naïve Bayes classification. K-Means clustering is used to cluster all data into the corresponding group based on data behavior, i.e. malicious and non-malicious, while the Naïve Bayes classifier is used to classify clustered data into correct categories, i.e. R2L, U2R, Probe, DoS and Normal. Experiments have been carried out to evaluate the performance of the proposed approach using KDD Cup ’99 dataset. The results showed that our proposed approach significantly improves the accuracy, detection rate up to 99.6% and 99.8%, respectively, while decreasing false alarms to 0.5%.

scholarly journals COMPATIBILITY OF SELECTION OF STUDENT DEPARTMENTS USING k-NEAREST NEIGHBOR AND NAÏVE BAYES CLASSIFIER IN INFORMATICS PRIVATE VOCATIONAL SCHOOL, SERANG CITY

2021 ◽  
Vol 4 (1) ◽  
pp. 33-39
Author(s):  
Budi Pangestu ◽  
Keyword(s):  
Nearest Neighbor ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Naive Bayes Classifier ◽  
Bayes Classifier ◽  
K Nearest Neighbor ◽  
Prospective Students ◽  
Naïve Bayes Classifier ◽  
K Nearest Neighbors ◽  
Selection Of

Selection of majors by prospective students when registering at a school, especially a Vocational High School, is very vulnerable because prospective students usually choose a major not because of their individual wishes. And because of the increasing emergence of new schools in cities and districts in each province in Indonesia, especially in the province of Banten. Problems experienced by prospective students when choosing the wrong department or not because of their desire, so that it has an unsatisfactory value or value in each semester fluctuates, especially in their Productive Lessons or Competencies. To provide a solution, a departmental suitability system is needed that can provide recommendations for specialization or major suitability based on students' abilities through attributes that can later assist students in the suitability of majors. The process of classifying the suitability of majors in data mining uses the k-Nearest Neighbor and Naive Bayes Classifier methods by entering 16 (sixteen) criteria or attributes which can later provide an assessment of students through this test when determining the majors for themselves, and there is no interference from people. another when choosing a major later. Research that has been carried out successfully using the k-Nearest Neighbors method has a higher recall of 99%, 81% accuracy and 82% precision compared to the Naïve Bayes Classifier whose recall only yields 98% while the accuracy and precision is the same as the k- Nearest Neighbors.

scholarly journals Perbandingan Seleksi Fitur Term Frequency & Tri-Gram Character Menggunakan Algoritma Naïve Bayes Classifier (Nbc) Pada Tweet Hashtag #2019gantipresiden

Kilat ◽  
2020 ◽  
Vol 9 (1) ◽  
pp. 103-114
Author(s):  
Arini - Arini ◽  
Luh Kesuma Wardhani ◽  
Dimas - Octaviano
Keyword(s):  
Feature Selection ◽  
Naive Bayes ◽  
High Accuracy ◽  
Naïve Bayes ◽  
Training Data ◽  
Naive Bayes Classifier ◽  
Bayes Classifier ◽  
Naïve Bayes Classifier ◽  
Term Frequency ◽  
Selection Of

Towards an election year (elections) in 2019 to come, many mass campaign conducted through social media networks one of them on twitter. One online campaign is very popular among the people of the current campaign with the hashtag #2019GantiPresiden. In studies sentiment analysis required hashtag 2019GantiPresiden classifier and the selection of robust functionality that mendaptkan high accuracy values. One of the classifier and feature selection algorithms are Naive Bayes classifier (NBC) with Tri-Gram feature selection Character & Term-Frequency which previous research has resulted in a fairly high accuracy. The purpose of this study was to determine the implementation of Algorithm Naive Bayes classifier (NBC) with each selection and compare features and get accurate results from Algorithm Naive Bayes classifier (NBC) with both the selection of the feature. The author uses the method of observation to collect data and do the simulation. By using the data of 1,000 tweets originating from hashtag # 2019GantiPresiden taken on 15 September 2018, the author divides into two categories: 950 tweets as training data and 50 tweets as test data where the labeling process using methods Lexicon Based sentiment. From this study showed Naïve Bayes classifier algorithm accuracy (NBC) with feature selection Character Tri-Gram by 76% and Term-Frequency by 74%,the result show that the feature selection Character Tri-Gram better than Term-Frequency.

scholarly journals An Enhanced Hybrid Intrusion Detection Mechanism Based on Chicken Swarm Optimization and Naïve-Bayes Method

2019 ◽  
Vol 8 (3) ◽  
pp. 5906-5910
Keyword(s):  
Intrusion Detection ◽  
Naive Bayes ◽  
Naïve Bayes ◽  
Naive Bayes Classifier ◽  
Bayes Classifier ◽  
Naïve Bayes Classifier ◽  
Detection Mechanism ◽  
Swarm Optimization ◽  
Network Intrusion ◽  
Chicken Swarm Optimization

The major important factor of network intrusion detection is to avoid malicious process in network. Since, existing modules are out-dated because of improper authentication and the network may get affected because of new attacks and malwares. In this research, Hybrid module is formed by using Chicken Swarm Optimization and Naive Bayes classifier (HCSONB) for classification of intrusion data. The hybrid method is introduced to detect the features efficiently in complex dataset because strategy which is designed to be capable of detecting huge data in network. Some traditional methods results in serious limitations in case of complex datasets. The algorithms are shared their properties together to discover better optimization results and the classification precisions values. This paper examines the feature selection performance by utilizing NSLKDD-99 dataset and comparing it with the Swarm Intelligence (SI), Naïve-Bayes classifier and proposed HCSO-NB algorithms. The proposed classification process designed in NETBEANS 8.2 tool. Experiments show that proposed HCSO-NB successfully improved the accuracy

scholarly journals KLASIFIKASI PESAN GANGGUAN PELANGGAN MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER

Kilat ◽  
2018 ◽  
Vol 7 (2) ◽  
pp. 100-108
Author(s):  
Haryono Haryono ◽  
Pritasari Palupiningsih ◽  
Yessy Asri ◽  
Andi Nikma Sri Handayani
Keyword(s):  
Naive Bayes ◽  
Naïve Bayes ◽  
Naive Bayes Classifier ◽  
Bayes Classifier ◽  
Naïve Bayes Classifier ◽  
Bayes Method ◽  
Machine Learning Methods ◽  
Data Processor ◽  
Selection Of

The application of customer disturbance message classifiers is made because of the process of reporting the interruption by the customer must be done by selection of data disorders by one by the admin to be able to follow-up from the existing customer reports. Naive Bayes is one of machine learning methods that uses probability calculations where the algorithm takes advantage of probability and statistical methods that predict future probabilities based on past experience. The application of the naive bayes classifier method with text mining as the initial data processor of the disorder messaging application can be concluded that this study yields an accuracy of probability values of 95 percent and proves that the Naive Bayes method can be used to help classify interference messages sent by customers.

Sign in / Sign up

Export Citation Format

Share Document