For people in developing countries, cassava is a major source of calories and carbohydrates. However, Cassava Mosaic Disease (CMD) has become a major cause of concern among farmers in sub-Saharan Africa countries, which rely on cassava for both business and local consumption. The article proposes a novel deep residual convolution neural network (DRNN) for CMD detection in cassava leaf images. With the aid of distinct block processing, we can counterbalance the imbalanced image dataset of the cassava diseases and increase the number of images available for training and testing. Moreover, we adjust low contrast using Gamma correction and decorrelation stretching to enhance the color separation of an image with significant band-to-band correlation. Experimental results demonstrate that using a balanced dataset of images increases the accuracy of classification. The proposed DRNN model outperforms the plain convolutional neural network (PCNN) by a significant margin of 9.25% on the Cassava Disease Dataset from Kaggle.

No abstract

No abstract

Machine Learning has found application in solving complex problems in different fields of human endeavors such as intelligent gaming, automated transportation, cyborg technology, environmental protection, enhanced health care, innovation in banking and home security, and smart homes. This research is motivated by the need to explore the global structure of machine learning to ascertain the level of bibliographic coupling, collaboration among research institutions, co-authorship network of countries, and sources coupling in publications on machine learning techniques. The Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) was applied to clustering prediction of authors dominance ranking in this paper. Publications related to machine learning were retrieved and extracted from the Dimensions database with no language restrictions. Bibliometrix was employed in computation and visualization to extract bibliographic information and perform a descriptive analysis. VOSviewer (version 1.6.16) tool was used to construct and visualize structure map of source coupling networks of researchers and co-authorship. About 10,814 research papers on machine learning published from 2010 to 2020 were retrieved for the research. Experimental results showed that the highest degree of betweenness centrality was obtained from cluster 3 with 153.86 from the University of California and Harvard University with 24.70. In cluster 1, the national university of Singapore has the highest degree betweenness of 91.72. Also, in cluster 5, the University of Cambridge (52.24) and imperial college London (4.52) having the highest betweenness centrality manifesting that he could control the collaborative relationship and that they possessed and controlled a large number of research resources. Findings revealed that this work has the potential to provide valuable guidance for new perspectives and future research work in the rapidly developing field of machine learning.

Synchronous and asynchronous e-learning are two popular e-learning modes that are commonly used in distant learning education. The study investigates how synchronous and asynchronous e-learning affect the academic performance of students. A questionnaire was used to collect data for this study from some students of the National Open University of Nigeria. The findings showed that students' attitude to synchronous and asynchronous e-learning affect their academic performance. The results demonstrated that only 60% of the respondents understand what asynchronous and synchronous e-learning means. Also, only 55% of the respondents believed that asynchronous and synchronous e-learning mode has a positive impact on their academic performance. Moreover, only 52% of the respondents are of the opinion that the curriculum in use at National Open University needs to be updated to increase the impact of the e-learning mode on the learners.

In sequential data, prediction and image classification, deep learning methods have obtained outstanding results. In this study, we propose image transformation of time series crude oil price by incorporating Directed Acyclic Graph to Convolutional Neural Network (DAG) based on image processing characteristics. Crude oil price time series is converted into 2-D images, utilizing 10 distinctive technical indicators. Geometric Brownian Motion was utilized to produces data for a 10-day time span. Thus, a 2-D image with a size of 10x10 sized 2-D is constructed. Then mark each image as "buy" or "sell" based on time series returns. The results show that integrating DAG with CNN improves the prediction accuracy by 14.18%. DAG perform best with an accuracy of 99.16%, sensitivity of 100% and specificity of 99.19%. COVID-19 has negatively affected Nigeria crude oil price which indicates a downward trend of crude oil price. The study recommends poly-cultural economy of Nigeria economy for national development of the nation.

Abstract-Breast cancer is one of the causes of female death in the world. Mammography is commonly used for distinguishing malignant tumors from benign ones. In this research, a mammographic diagnostic method is presented for breast cancer biopsy outcome predictions using five machine learning which includes: Logistic Regression (LR), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Random Forest (RF) and Support Vector Machine (SVM) classification. The testing results showed that SVM learning classification performs better than other with accuracy of 95.8% in diagnosing both malignant and benign breast cancer, a sensitivity of 98.4% in diagnosing disease, a specificity of 94.4%. Furthermore, an estimated area of the receiver operating characteristic (ROC) curve analysis for Support vector machine (SVM) was 99.9% for breast cancer outcome predictions, outperformed the diagnostic accuracies of Logistic Regression (LR), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Random Forest (RF) methods. Therefore, Support Vector Machine (SVM) learning classification with mammography can provide highly accurate and consistent diagnoses in distinguishing malignant and benign cases for breast cancer predictions.

No abstract