Transactions on Machine Learning and Artificial Intelligence <p>Transactions on Machine Learning and Artificial Intelligence is peer-reviewed open access online journal that provides a medium of the rapid publication of original research papers, review articles, book reviews and short communications covering all areas of machine learning and artificial Intelligence. The journal publishes state-of-the-art research reports and critical evaluations of applications, techniques and algorithms in machine learning, artificial intelligence, cognitive science, software engineering, database systems, soft computing, optimization and modelling and related application areas.</p> en-US (Thomas Harvey) (Olivia Adam) Fri, 30 Aug 2019 00:00:00 +0000 OJS 60 A Population-Based Multicriteria Algorithm for Alternative Generation <p>Complex problems are frequently overwhelmed by inconsistent performance requirements and incompatible specifications that can be difficult to identify at the time of problem formulation. Consequently, it is often beneficial to construct a set of different options that provide distinct approaches to the problem. These alternatives need to be close-to-optimal with respect to the specified objective(s), but be maximally different from each other in the solution domain. The approach for creating maximally different solution sets is referred to as modelling-to-generate-alternatives (MGA). This paper introduces a computationally efficient, population-based multicriteria MGA algorithm for generating sets of maximally different alternatives.</p> Julian Scott Yeomans Copyright (c) 2019 Transactions on Machine Learning and Artificial Intelligence Sun, 08 Sep 2019 08:02:00 +0000 Time Series Analysis of Nigeria Foreign Exchange Reserve Time series analysis was carried out on Nigeria External Reserves for the period of 1960 – 2018. An empirical investigation was conducted using time series data on Nigeria External Reserve for a period of 58 years. The techniques of estimation employed in the study include Phillips-Perron unit root test, Dickey-Fuller’s test, the Autocorrelation function and the Partial Autocorrelation function for the model selection. The Box-Jenkins ARIMA methodology was used for forecasting the monthly data collected from 1960 to 2018. Result of the analysis revealed that the series became stationary at first difference. The diagnostic check showed that ARIMA (1, 1, 2) is appropriate or optimal model based on the Loglikelihood (LogLik), Akaike’s Information Criterion (AIC), as well as the small standard error of the AR(1), MA(1) and MA(2) parameters. The performance of “forecast.Arima()” function in R gives the best model for Nigeria external reserve. Testing for other ARIMA models is necessary in order to establish the best. The downward movement in the forecasts of Nigeria external reserve would be helpful for policy makers in Nigeria. Isaac O. Ajao Copyright (c) 2019 Transactions on Machine Learning and Artificial Intelligence Sun, 08 Sep 2019 08:01:24 +0000 The Theory Graph Modeling and Programming Paradigm Systems FROM Modules TO the Application Areas <h2>The mathematical basics of graph modeling and paradigm programming of applied systems (AS) are presented. The vertices of graph are been &nbsp;the functional elements of the systems and the arcs define the connections between them. The graph is represented by an adjacency and reach ability matrix. A number of graph of program structures and their representation by mathematical operations (unions, connections, differences, etc.) are shown. Given the characteristics of graph structures, complexes, units, and systems created from the modules of the graph. The method of modelling the system on the graph of&nbsp; modules, which describe &nbsp;in the&nbsp; programming languages – LP (Algol-60, Fortran, Cobol, PL/1, Smalltalk, etc.) and the advanced operations of association (assembling, make,&nbsp; weaver, config &nbsp;etc.). The standard of configuration (2012) Assembly of heterogeneous software elements in AS of different fields of knowledge is made. A brief description of modern and future programming paradigms for formal theoretical creation of systems from intelligent and cloud service elements of the Internet is given.. There is a new direction of nanotechnology in the near future.</h2> E.M. Lavrischeva Copyright (c) 2019 Transactions on Machine Learning and Artificial Intelligence Sun, 08 Sep 2019 08:02:37 +0000 I-AFYA: INTELLIGENT SYSTEM FOR THE MANAGEMENT OF DIABETES IN KENYA. <p>Computational Intelligence approaches have gained increasing popularity given their ability to cope with large amounts of clinical data and uncertain information. The treatment offered for diabetes aims to keep a patients' blood glucose level as normal as possible and to prevent health complications developing later in their life. Researchers and developers have created diabetes applications and systems that already are frequent on various application stores and shelves. Applications running on&nbsp;artificial intelligence&nbsp;(AI) and cognitive computing models offer promise in diabetes&nbsp;care. This is given the fact that diabetes is a global pandemic. An estimated 425 million people worldwide have diabetes, accounting for 12% of the world's health expenditures and yet one in two persons remain undiagnosed and untreated. Type 2 diabetes is driven by the global obesity epidemic and a sedentary lifestyle that overwhelms the body's internal glucose control requiring exogenous insulin. In Kenya alone, diabetes is a leading cause of kidney failure, lower limb amputations and adult-onset blindness. Thus, research on diabetes care using technological (ICT) solutions will continue to dominate the discussion for quite some time.&nbsp; The early detection of diabetes is of paramount importance. Generally, a physician diagnoses diabetes by evaluating the current test results of a patient or by comparing the patient with other patients who have the same condition. The early detection and screening for individuals with impaired glucose tolerance can help lower risk of developing diabetes and reduce the long-term burden to individuals and health services. For this reason, artificial intelligent systems for diagnosing diabetes have been an item for research for some time. The use of intelligent systems in the Kenyan health care system can help lower the cost of diabetes treatment besides increasing the access and quality of health care provided to diabetic patients.&nbsp;</p> Nzioka Nguku Copyright (c) 2019 Transactions on Machine Learning and Artificial Intelligence Sun, 08 Sep 2019 08:03:11 +0000