Browsing by Author "Sadiku, Peter Ogirima"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
Item An Ensemble Approach Based on Decision Tree and Bayesian Network for Intrusion Detection(Computers and Applied Computer Science Faculty in "Tibiscus" University of Timişoara, Romania., 2017-06-01) Balogun, Abdullateef Oluwagbemiga; Balogun, Adedayo Miftaudeen; Sadiku, Peter Ogirima; Amusa, LateefThis paper presents an overview of intrusion detection and a hybrid classification algorithm based on ensemble method (stacking) which uses decision tree (J48) and Bayesian network as base classifiers and functional tree algorithm as the meta-learner. The data set is passed through the decision tree and node Bayesian network for classification. The meta-learner (Functional tree classifier) will then select the value of the base classifier that has the higher accuracy based on majority voting. The key idea here is to always pick the value with higher accuracy since both base classifier (decision tree and Bayesian network) will always classify all instances. A performance evaluation was performed using a 10-fold cross validation technique on the individual base classifiers (decision tree and Bayesian network) and the ensemble classifier (DT-BN) using the KDD Cup 1999 dataset on WEKA tool. Experimental results show that the hybrid classifier (DT-BN) gives the best result in terms of accuracy and efficiency compared with the individual base classifiers (decision tree and BN). The decision tree gave a result of (99.9974% for DoS, 100% for Normal, 98.8069% for probing, 97.6021% for U2R and 73.0769% for R2L), the Bayesian network (99.6410% for DoS, 100% for Normal, 97.1756% for probing, 97.0693% for U2R and 69.2308% for R2L),while the ensemble method gave a result of (99.9977% for DoS, 100% for Normal, 98.8069% for probing, 97.6909% for U2R and 73.0769% for R2L).Item Heterogeneous Ensemble Models For Generic Classification(Computers and Applied Computer Science Faculty in "Tibiscus" University of Timişoara, Romania., 2017-05-10) Balogun, Abdullateef Oluwagbemiga; Balogun, Adedayo Miftaudeen; Sadiku, Peter Ogirima; Adeyemo, Victor ElijahThis paper presents the application of somedata mining techniques in the field of health care and computer network security. The selected classifiers wer eused individually and also, they were ensemble methods using four different combinations for the purpose of classification. Naïve Bayes, Radial Basis Function and Ripper algorithms were selected and the ensemble methods were majority voting, multi-scheme, stacking and Minimum Probability. The KDDCup’99 dataset was used as the benchmark for computer network security, while for the health care, breast cancer and diabetes dataset from the WEKA repository were used. All experiments and simulations were carried out, analyzed and evaluated using the WEKA tool. The Multi-scheme ensemble method gave the best accuracy result for the KDD dataset (99.81%) and the breast cancer dataset (73.08%) but its value of (75.65%) on breast cancer is the least of them all. Ripper algorithm gave the best result accuracy (99.76%) on KDD dataset amongst the base classifier but it was slightly behind in the breast cancer and diabetes dataset.Item Multiple Ceaser Cipher Encryption Algorithm(Mathematical Association of Nigeria (MAN)., 2017-12) Balogun, Abdullateef Oluwagbemiga; Sadiku, Peter Ogirima; Mojeed, Hameed Adeleye; Raifu, Hameed AdetunjiThe Caesar cipher has always been the major reference point when cryptographic algorithms (also called ciphers) are discussed. This, probably, is due to its being an age-long cipher. It may also be owing to the belief that the Caesar cipher was the first cipher used ever. Caesar cipher operation is based on shift-by-3 rule which makes its breaking obviously easy since an exhaustive key search of the other 25 keys can be conveniently performed. Ipso facto, an investigation into an enhancement of this too-simple-to-crack cipher is invariably necessary and ultimately important. This study is, therefore, concerned with developing a new enhanced model of Caesar cipher for a better security using multiple encryption technique, whereby an already-encrypted message is encrypted one or more times using the same or different algorithm. The new model works by wrapping a plaintext message in three crypto-wrappers and each encryption/decryption phase uses a different shift key from the other. The model supports both uppercase and lowercase characters. However, the model does not encrypt/decrypt numbers, special characters, whitespace, and file types such as word document, binary, or pdf files, but only text files. Most importantly, the new enhanced model is able to provide a better security of message by encrypting a plaintext message three times; in this way, brute forcing or an exhaustive key search will be difficult to perform; thus, making cryptanalysis almost a mirage!Item A Network Intrusion Detection System: Enhanced Classification via Clustering Model(Research Nexus Africa’s Networks in Conjunction with The African Institute of Development Informatics & Policy (AIDIP) Ghana & The International Centre for Information Technology & Development (ICITD), USA, 2015-12-10) Balogun, Abdullateef Oluwagbemiga; Balogun, Adedayo Miftaudeen; Adeyemo, Victor Elijah; Sadiku, Peter OgirimaThe aim of developing an IDS is to build a system that oversee the general protection of a network from attacks both from withinand without, and doing so accurately. Optimization of IDS has also being receiving attention from the research community due toits large volumes of security audit data. In developing an IDS, most dataset used have high dimension in which only few attributes are needed for building an IDS – feature selection is used to solve this little problem. In this paper, we present and analyze the performance of some machine learning algorithm which performs classification via clustering using the KDDcup’99 dataset. Using the WEKA tool, simulations were ran and results was deduced after applying the proposed models to the dataset containing all the type of attacks