DSpace Angular :: Browsing by Author "Dauda K. A."

Browsing by Author "Dauda K. A."

Now showing 1 - 7 of 7

A Novel Hybrid Dimension Reduction Technique for Efficient Selection of Bio-marker Genes and Prediction of Heart Failure Status of Patients
(Elsevier, 2021) Dauda K. A.; Olorede K. O.; Aderoju S. A.
This study highlighted and provided a conceptual framework of a hybridized dimension reduction by combining Genetic Algorithms (GA) and Boruta Algorithm (BA) with Deep Neural Network (DNN). Among questions left unanswered sufficiently by both computational and biological scientists are: which genes among thousand of genes are statistically relevant to the prediction of patients’ heart rhythm? and how they are associated with heart rhythm? A plethora of models has been proposed to reliably identify core informative genes. The premise of this present work is to address these limitations. Five distinct micro-array data on heart diseases have been taken into consideration to observe the prominent genes. We form three distinct set two-way hybrids between Boruta Algorithm and Neural Network (BANN); Genetic Algorithm and Deep Neural Network (GADNN) and Boruta Algorithm and Deep Neural Network (BADNN), respectively, to extract highly differentially expressed genes to achieve both better estimation and clearer interpretation of the parameters included in these models. The results of the filtering process were observed to be impressive since the technique removed noisy genes. The proposed BA algorithm was observed to select minimum genes in the wrapper process with about 80% of the five datasets than the proposed GA algorithm with 20%. Moreover, the empirical comparative results revealed that BADNN outperformed other proposed algorithms with prediction accuracy of 97%, 87%, and 100% respectively. Finally, this study has successfully demonstrated the utility, versatility, and applicability of hybrid dimension reduction algorithms (HDRA) in the realm of deep neural networks.
A Test Procedure for Ordered Hypothesis of Population Proportions Against a Control
(Turkiye Klinikleri Journal of Biostatistics, 2016) Yahya W. B.; Olaniran O. R.; Garba M. K.; Oloyede I.; Banjoko A. W.; Dauda K. A.; Olorede K. O.
Objective: This paper aims to present a novel procedure for testing a set of population proportions against an ordered alternative with a control. Material and Methods: The distribution of the test statistic for the proposed test was determined theoretically and through Monte-Carlo experiments. The efficiency of the proposed test method was compared with the classical Chi-square test of homogeneity of population proportions using their empirical Type I error rates and powers at various sample sizes. Results: The new test statistic that was developed for testing a set of population proportions against an ordered alternative with a control was found to have a Chi-square distribution with non-integer values degrees of freedom v that depend on the number of population groups k being compared. Table of values of v for comparing up to 26 population groups was constructed while an expression was developed to determine v for cases where k > 26. Further results showed that the new test method is capable of detecting the superiority of a treatment, for instance a new drug type, over some of the existing ones in situations where only the qualitative data on users' preferences of all the available treatments (drug types) are available. The new test method was found to be relatively more powerful and consistent at estimating the nominal Type I error rates (α), especially at smaller sample sizes than the classical Chi-square test of homogeneity of population proportions. Conclusion: Conclusion: The new test method proposed here could find applications in pharmacology where a newly developed drug might be expected to be more preferred by users than some of the existing ones. This kind of test problem can equally exist in medicine, engineering and humanities in situations where only the qualitative data on users' preferences of a set of treatments or systems are available.
Effectiveness of Contraceptive Usage among Reproductive Ages in Nigeria Using Artificial Neural Network (ANN)
(University of the West of Scotland, School of Computing, 2018) Dauda K. A.; Babatunde A. N.; Olorede K. O.; Abdulsalam S. O.; Ogundokun O. R.
Purpose: Contraceptive usage among women of reproductive age is a fertile area of research for the medical scientists, social scientists, and medical practitioners. However, not much work has been done in Nigeria to identify some of the contraceptive methods used in preventing unwanted pregnancy and sexually transmitted infection (STI) diseases, particularly to explicitly determine which of them are most effective. This paper tackles this task. Methodology: In this paper, we examine different types of contraceptive methods used by Nigeria women to prevent unwanted pregnancy. Datasets on contraceptive usage among 28,647 women of reproductive age from the 2008 Nigerian Demographic Health Survey (NDHS) were analyzed using Artificial Neural Network (ANN) approach in the R language. Finding/Results: The results of the analysis revealed that approximately 1%, 3% and 9% of the study population were using Folkloric, Traditional and Modern methods, respectively. Additionally, 12%, 3% and 9% of the women were found to be currently using the methods since last birth and before last birth. Furthermore, the ANN results through the Garson algorithm revealed that using the modern methods (IUD, Norplant and Female condom) and the traditional methods prevents the unwanted pregnancy negatively, while other modern methods (pill, male condom, female sterilizer, periodic abstinence, withdrawal, locational amenorrhea and foam or jelly) were found to be effective in preventing unwanted pregnancy positively. Interestingly, women that used the modern methods (i.e. Pill, IUD, Injection, Male condom, Female sterilization, Periodic abstinence, Withdrawal and Lactational amenorrhea) were found to be effective in preventing unwanted pregnancy. Research Limitation: The main limitation of this study is the inability to access the current data from NDHS. Therefore, the conclusion of this study is only based on the 2008 NDHS data. Originality/Value: This research work introduces artificial neural network (ANN) to determine and identify the effect of contraceptive methods usage among Nigeria women, and to determine their level of important using Garson's algorithm variable important. This introduction measures the nonlinear effect that exists between the methods and response variable (pregnant and no pregnant) which existing research approaches do not. Keywords: Contraceptive methods, Variable Important, Garson Algorithm, Artificial Neural Network (ANN) and Data mining.
Efficient Support Vector Machine Classification of Diffused Large B-Cell Lymphoma and Follicular Lymphoma mRNA Tissue Samples
(Annals Computer Science Series, 2015) 11. Banjoko A. W.; Yahya W. B.; Garba M. K.; Olaniran O. R.; Dauda K. A.; Olorede K. O.
In this study, an efficient Support Vector Machine (SVM) algorithm that incorporates feature selection procedure for efficient identification and selection of gene biomarkers that are predictive of Diffuse Large B–Cell Lymphoma (DLBCL) and Follicular Lymphoma (FL) cancer tumor samples is presented. The data employed were published real life microarray cancer data that contained 7,129 gene expression profiles measured on 77 biological samples that comprised 58 DLBCL and 19 FL tissue samples. The dimension reduction approach of the Welch statistic was employed at the feature selection phase of the SVM algorithm. The cost and kernel parameters of the SVM model were tuned over a 10–fold cross-validation to improve the efficiency of the SVM classifier. The entire sample was randomly partitioned into 95% training and 5% test samples. The SVM classifier was trained using Monte Carlo Cross validation approach with 1000 replications. The performance of this classifier was assessed on the test samples using misclassification error rate (MER) and other performance measures. The results showed that the SVM classifier is quite efficient by yielding very high prediction accuracy of the tumor samples with fewer differentially expressed genes. The selected gene biomarkers in this work can be subjected to further clinical screening for proper determination of their biological relationship with DLBCL and FL tumour sub groups. However, more studies with large samples might be needed in future to validate the results from this work.
Multiclass Feature Selection and Classification with Support Vector Machine in Genomic Study
(Professional Statisticians Society of Nigeria (PSSN), 2017) Banjoko A. W.; Yahya W. B.; Garba M. K.; Olaniran O. R.; Amusa L. B.; Gatta N. F.; Dauda K. A.; Olorede K. O.
This study proposes an efficient Support Vector Machine (SVM) algorithm for feature selection and classification of multiclass response group in high dimensional (microarray) data. The Feature selection stage of the algorithm employed the F-statistic of the ANOVA–like testing scheme at some chosen family-wise-error-rate (FWER) to control for the detection of some false positive features. In a 10-fold cross validation, the hyper-parameters of the SVM were tuned to determine the appropriate kernel using one-versus-all approach. The entire simulated dataset was randomly partitioned into 95% training and 5% test sets with the SVM classifier built on the training sets while its prediction accuracy on the response class was assessed on the test sets over 1000 Monte-Carlo cross-validation (MCCV) runs. The classification results of the proposed classifier were assessed using the Misclassification Error Rates (MERs) and other performance indices. Results from the Monte-Carlo study showed that the proposed SVM classifier was quite efficient by yielding high prediction accuracy of the response groups with fewer differentially expressed features than when all the features were employed for classification. The performance of this new method on some published cancer data sets shall be examined vis-à-vis other state-of-the-earth machine learning methods in future works.
On the Efficiency of Maximum Likelihood and Restricted Maximum Likelihood Estimators for Linear Mixed Effects Models
(Mathematical Association of Nigeria (MAN), 2016) 3. Yahya W. B.; Adeniyi I. A.; Olayemi R. M.; Dauda K. A.; Banjoko, A. W.; Olorede K. O.; Ezenweke C. P.
In this work, the performances of maximum likelihood (ML) and restricted maximum likelihood (REML) estimation of linear mixed effects models were examined. The behaviours and properties of the ML and REML methods of estimating the variance-covariance components in linear mixed effects models were studied via Monte-Carlo experiment. Results revealed that both estimation techniques yielded the same estimates of the fixed effects parameters. However, ML estimates of variance-covariance components, especially for the random effects terms were more biased downwards than the REML counterparts. It was also discovered that the biasness of the ML estimates for the variance-covariance components increase as the number of fixed effects parameters in the model increase.
Partial Least Squares-Based Classification and Selection of Predictive Variables of Crimes against Properties in Nigeria
(Professional Statisticians Society of Nigeria (PSSN), 2017) Olorede K. O.; Yahya W. B.; Garuba A. O.; Banjoko A. W.; Dauda K. A.
In this study, the state-of-the-art Partial Least Squares (PLS) based models (PLS-Discriminant analysis (PLS-DA), Sparse PLS-DA (SPLS-DA) and Sparse Generalized PLS (SGPLS)) were employed to model and classify the rate of crimes (low or high) committed against properties across the 36 states in Nigeria and the Federal Capital Territory (FCT). The core variables that are predictive of this crime type in Nigeria were identified using the LASSO penalty method via the PLS. Data on occurrences of cases of offences against property obtained from the data base of Nigerian Police Force were utilized in this study. The missing values due to non-occurrence or non-reportage of crime cases were imputed, using the techniques of multivariate imputation by chained equation. The complete data set were partitioned into training and test sets using 80:20 holdout scheme. The 80% training set was used to build the PLS-based models that were in turn used to predict the overall crime rates of Nigerian cities in the 20% held out test data over 200 Monte-Carlo cross-validation runs. All the PLS-based models yielded good classification of unseen test samples into either of two qualitative classes of high and low crime rates with average Correct Classification Rate (CCR) of 94%. Other performance metrics including sensitivity, specificity, positive and negative predictive values, balance accuracy and diagnostic odds ratio were estimated to further examine their classification efficiencies. The SGPLS identified fewer (just 3 out of 12) core relevant crime variables that are predictive of the overall crime rates in Nigerian states with highest CCR than the SPLS which selected 9 such variables to achieved about the same feat.

Browsing by Author "Dauda K. A."

Results Per Page

Sort Options