Browsing by Author "Mabayoje, Modinat Abolore"
Now showing 1 - 11 of 11
Results Per Page
Sort Options
Item Comparative Analysis of Selected Supervised Classification Algorithms.(Computer Chapter of the Institute of Electrical & Electronics Engineers (IEEE) Nigeria Section., 2015-10) Mabayoje, Modinat Abolore; Balogun, Abdullateef Oluwagbemiga; Salihu, Shakirat Aderonke; Oladipupo, Kehinde RazakInformation is not packaged in a standard easy-to-retrieve format. It is an underlying and usually subtle and misleading concept buried in massive amounts of raw data. From the beginning of time it has been man’s common goal to make his life easier. The prevailing notion in society is that wealth brings comfort and luxury, so it is not surprising that there has been so much work done on ways to sort large volume of data. Over the year, there are various data mining techniques and used to sort large volume of data. This paper considers Classification which is a supervised learning technique. Therefore the need to come up with the most efficient way to deal with voluminous data with very little time frame has been one of the biggest challenges to the AI community. Hence, this paper presents a comparative analysis of three classification algorithms namely; Decision Tree (J-48), Random Forest and Naïve Bayes. A 10-fold cross validation technique is used for the performance evaluation of the classifiers on KDD’’99, VOTE and CREDIT datasets using WEKA (Waikato Environment forKnowledge Analysis) tool. The experiment shows that the type of dataset determines which classifier is suitable.Item Email Data Security Using Cryptography(Faculty of Communication and Information Sciences, University of Ilorin, Ilorin, Nigeria., 2016) Bajeh, Amos Orenyi; Ayeni, A .O.; Balogun, Abdullateef Oluwagbemiga; Mabayoje, Modinat AboloreEmail has become one of the prominent medium of message transmission between web users. Thus, email security measures need to evolve from time to time in order to mitigate security threats which are also evolving and becoming sophisticated. This paper presents a study that proposed, implemented and evaluated the performance of an email security approach that utilizes the Rivest Shamir Adleman algorithm as an additional layer of security. The approach involves an application that converts messages into cipher text as a separate platform.The cipher text is transmitted as the original message on the email platform. Also, the cryptographic keys are transmitted between communicating users via another media such as mobile phone.The proposed approach is evaluated by measuring the accuracy of the message transmitted between users. It showed an average accuracy of 98% overall the scenarios examined.Item Enhanced Classification via Clustering Techniques using Decision Tree for Feature Selection(Foundation of Computer Science (FCS), NY, USA, 2015-09-01) Balogun, Abdullateef Oluwagbemiga; Mabayoje, Modinat Abolore; Salihu, Shakirat Aderonke; Salvation, ArinzeInformation overload has raggedly increased as a result of the advances in the aspect of storage capabilities and data collection in previous years. The growth seen in the number of observation has partly cause a collapse in analytical method but the increases in the number of variable associated with each observation has grossly collapse it. The number of variables that are measured on each observation.is referred to as the dimension of the data, and a major problem of dataset containing high dimensions is that, there exist only few “important” measured variables for understanding the fundamental occurrences of interest. Hence, dimension reduction of the original data prior to any modeling of the data is of great necessity today. In this paper, a précis of K-Means, Expectation Maximization and J48 decision tree classifier is presented with a framework on the performance measurement of base classifiers with and without feature reduction. A performance evaluation was carried out based on F-Measure, Precision, Recall, True Positive Rate, False Positive Rate, ROC Area and Time taken to build model. The experiment revealed that the reduced dataset yielded improved results than the full dataset after performing classification via clustering.Item A Framework for Coordinating Usability Engineering and Software Engineering Activities in the Development of Content Management Systems(AIMS Research Journal Publication Series The International Centre for Information Technology & Development (ICITD), USA., 2017-06) Balogun, Abdullateef Oluwagbemiga; Mabayoje, Modinat Abolore; Adeniyi, Ebenezer Olufemi; Salihu, Shakirat AderonkeDue to the expansion of the internet in recent years, we have witnessed an increase in popularity of web applications and its technologies. A particular technology of web application, Content Management system, has also gained relevance as they facilitate the distribution of wide varieties of content. The process involved in designing and developing content management system is a complex procedure due to the variability of its requirement over time which has effects on its architecture and design. Currently, the Usability Engineering (UE) and Software Engineering (SE) processes are practiced as being independent of each other. However, several dependencies and constraints exist between these two frameworks, which make coordination between the UE and the SE teams crucial. Failure of coordination between the UE and SE teams leads to CMS that often lacks necessary functionality and impedes user performance. At the same time, the UE and SE processes cannot be integrated because of the differences in focus, techniques, and terminology. We therefore propose a development framework that incorporates SE and UE efforts to guide current CMS development. The framework characterizes the information exchange that must exist between the UE and SE teams during CMS development to form the basis of the coordinated development framework. The UE Scenario-Based Design (SBD) process provides the basis for identifying UE activities. Similarly, the Requirements Generation Model (RGM), and Structured Analysis and Design are used to identify SE activities. We identify UE and SE activities that can influence each other, and identify the high-level exchange of information that must exist among these activities. We further examine these interactions to gain a more in-depth understanding as to the precise exchange of information that must exist among them. The identification of interacting activities forms the basis of a coordinated development framework that incorporates and synchronizes the UE and SE processes.Item Gain Ratio and Decision Tree Classifier for Intrusion Detection(Foundation of Computer Science (FCS), NY, USA, 2015) Mabayoje, Modinat Abolore; Akintola, Abimbola Ganiyat; Balogun, Abdullateef Oluwagbemiga; Ayilara, OpeyemiWith the evident need for accuracy in the performance of intrusion detection system, it is expedient that in addition to the algorithms used, more activities should be carried out to improve accuracy and reduce real time used in detection. This paper reviews how data mining relates to IDS, feature selection and classification. This paper proposes architecture of IDS where GainRatio is used for feature selection and decision tree for classification using NSL-KDD99 dataset, It also includes the evaluation of the performance of the Decision tree on the dataset and also on the reduced dataset.Item Influence of Feature Selection On Multi-Layer Perceptron Classifier for Intrusion Detection System(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., 2016-12-15) Mabayoje, Modinat Abolore; Balogun, Abdullateef Oluwagbemiga; Ameen, Ahmed Oloduowo; Adeyemo, Victor ElijahThe usage of the most popular neural network – Multilayer perceptron, as gained ground for the purpose of detecting intrusion. A lot of researchers had used it judiciously but there exist problem of slow training time and data over-fitting. This paper reviews the various data mining techniques for applied in the area intrusion detection, categories of attacks, and techniques for feature selection. This paper proposes an architecture where information gain is used for feature selection and multilayer perceptron (MLP) for classification on KDD’99 dataset. Evaluation of the performance of the MLP classifier on the KDD’99 dataset and also on the reduced dataset was conducted.Item Parameter tuning in KNN for software defect prediction: an empirical analysis(Department of Computer Engineering, Universitas Diponegoro, Indonesia., 2019-10-31) Mabayoje, Modinat Abolore; Balogun, Abdullateef Oluwagbemiga; Jibril, Hajarah Afor; Atoyebi, Jelili Olaniyi; Mojeed, Hammed Adeleye; Adeyemo, Victor ElijahSoftware Defect Prediction (SDP) provides insights that can help software teams to allocate their limited resources in developing software systems. It predicts likely defective modules and helps avoid pitfalls that are associated with such modules. However, these insights may be inaccurate and unreliable if parameters of SDP models are not taken into consideration. In this study, the effect of parameter tuning on the k nearest neighbor (k-NN) in SDP was investigated. More specifically, the impact of varying and selecting optimal k value, the influence of distance weighting and the impact of distance functions on k-NN. An experiment was designed to investigate this problem in SDP over 6 software defect datasets. The experimental results revealed that k value should be greater than 1 (default) as the average RMSE values of k-NN when k>1(0.2727) is less than when k=1(default) (0.3296). In addition, the predictive performance of k-NN with distance weighing improved by 8.82% and 1.7% based on AUC and accuracy respectively. In terms of the distance function, kNN models based on Dilca distance function performed better than the Euclidean distance function (default distance function). Hence, we conclude that parameter tuning has a positive effect on the predictive performance of k-NN in SDP.Item PERFORMANCE EVALUATION OF SELECT DATA MINING SOFTWARE TOOLS FOR DATA CLUSTERING(Federal University Wukari, Taraba State, Nigeria., 2018-09-10) Ameen, Ahmed Oloduowo; Bajeh, Amos Orenyi; Adesiji, Boluwatife Aderinsola; Balogun, Abdullateef Oluwagbemiga; Mabayoje, Modinat AboloreData mining is used to discover knowledge from information system. Clustering is one of the techniques used for data mining. It can be defined as a technique of grouping un-labelled data objects such that objects belonging to one cluster are not similar to the objects belonging to another cluster. Data mining tools refer to the software that are used for the process of efficiently analysing, summarizing and extracting useful information from different perspectives of data. This paper presents a comparative analysis of four open-source data mining software tools (WEKA, KNIME, Tanagra and Orange) in the context of data clustering, specifically K-Means and Hierarchical clustering methods. The results of the performance analysis based on the execution time and quality of clusters showed that WEKA tool outperforms the other tools with the lowest SSE of 199.7308 with an average execution time of 1.535 seconds. Knime has SSE of 222.217 but with an average execution time of 7.13 seconds, and then Tanagra with SSE of 269.3902 and average execution time of 2.01 seconds, Orange has the poorest performance with SSE of 388.78.Item SOFTWARE DEFECT PREDICTION: EFFECT OF FEATURE SELECTION AND ENSEMBLE METHODS(Federal University Wukari, Taraba State, Nigeria., 2018-09-10) Mabayoje, Modinat Abolore; Balogun, Abdullateef Oluwagbemiga; Bajeh, Amos Orenyi; Musa, Badamasi AbubakarSoftware defect prediction is the process of locating defective modules in software. It facilitates testing efficiency and consequently software quality. It enables a timely identification of fault-prone modules. The use of single classifiers and ensembles for predicting defects in software has been met with inconsistent results. Previous analysis say ensemble are often more accurate and are less affected by noise in datasets, also achieving lower average error rates than any of the constituent classifiers. However, inconsistencies exist in these various experiments and the performance of learning algorithms may vary using different performance measures and under different circumstances. Therefore, more research is needed to evaluate the performance of ensemble algorithms in software defect prediction. Adding feature selection reduces data sets with fewer features and improves the classifiers and ensemble performance over the datasets. The goal of this paper is to assess the efficiency of ensemble methods in software defect prediction using feature selection. This study compares the performance of four ensemble algorithms using 11 different performance metrics over 11 software defect datasets from the NASA MDP repository. The results indicate that feature selection and use of ensemble methods can improve the classification results of software defect prediction. Bagged ensemble models have the best results. In addition, Voting and Stacking also performed better than individual base classifiers. In terms of single classifier, SMO performs best as it outperformed Decision Tree (J48), MLP, and KNN with and without feature selection. Thus, it can be derived that feature selection can help improve the accuracy of both individual classifiers and ensemble methods by removing noisy and inconsistent features in the datasets.Item Solving the Next Release Problem using a Hybrid Metaheuristic(Computers and Applied Computer Science Faculty in "Tibiscus" University of Timişoara, Romania., 2016) Balogun, Abdullateef Oluwagbemiga; Mabayoje, Modinat Abolore; Makinwa, Sayo Michael; Bajeh, Amos OrenyiThe Next Release Problem is characterized by the need to determine the features that are to be included in a particular software system to make up the next release. These features are to be selected, such that users’ demands and needs are satisfied as much as possible, given a limited resources, by ensuring that the available resources are used to develop the most important features first. This work applies a hybrid of Variable Neighbourhood Search (VNS) and Tabu Search (TS) for solving bi-objective NRP, using a cost-value model for requirements. Experiments showed the hybrid metaheuristics to produce a Pareto optimal set with a controllable dynamic number of options whose score and cost value range can be controlled via parameters that can be modified without a significant effect on execution time.Item Wrapper Feature Selection based Heterogeneous Classifiers for Software Defect Prediction(Adeleke University, Ede., 2019-02) Mabayoje, Modinat Abolore; Balogun, Abdullateef Oluwagbemiga; Bello, Shade Musllimat; Atoyebi, Jelili Olaniyi; Mojeed, Hammed Adeleye; Ekundayo, Ayobami HalimatThe performance of Software Defect Prediction (SDP) models depends on the quality of dataset used for training the models. The high dimensionality of software metric features has been noted as a data quality problem which affects the performance of SDP models. This makes it crucial to apply feature selection (FS) to SDP since FS can remove irrelevant and redundant software metric features. In this study, the effect of wrapper-based FS methods on classification techniques in SDP was investigated. The wrapper FS methods were based on different search methods; Best First Search (BFS), Genetic Search (GS), Greedy Stepwise Search (GSS) and Multi-Objective Evolutionary Search (MOES) so as to investigate their respective effect on classifiers in SDP. Five (5) publicly available software defect datasets were used. These datasets were classified by the individual classifiers which were carefully selected based on their characteristics hence the heterogeneity. Naïve Bayes (NB) was selected from Bayes category Classifier, K-Nearest Neighbor (KNN) was selected from Instance-Based Learner category and (J48) Decision Tree from Trees Function classifier. The experimental results clearly showed that the application of wrapper FS method to datasets before classification in SDP is better and should be encouraged as NB with GS based Wrapper Method had the best accuracy performance. It can be concluded that FS methods are capable of improving the performance of predictive models in SDP.