Executive SummaryEvery year, the Joint Admission Board (JAB) is tasked to determine those students who are expected to join various Kenyan public universities under the government sponsorship scheme. This exercise is usually extensive because of the large number of qualified students compared to the very limited number of slots at various institutions and the shortage of funding from the government. Further, this is made complex by the fact that the selections are done against a predefined cluster subjects vis a vis the student's preferred and applied for academic courses. Minimum requirements exist for each course and only students having the prescribed grades in specific subjects are eligible to join that course. Due to this, students are often admitted to courses they consider irrelevant to their career prospects and not their preferred choices.This process is tiresome, costly, and prone to bias, errors, or favour, leading to disadvantaging innocent students. This paper examines the potential use of artificial neural networks at the JAB for the process of selecting students for university courses. Based on the fact that Artificial Neural Networks (ANNs) have been tested and used in classification, the paper explains how a trained neural network can be used to perform the students' placement effectively and efficiently. JAB will be able, therefore, to undertake the students' placement thoroughly and be able to accomplish it with minimal wastage of time and resources respectively without having to utilise unnecessary effort. The paper outlines how the various metrics can be coded and used as input to the ANNs. Ultimately, the paper underscores the various merits that would accompany the adoption of this technique. By making use of neural networks in the university career choices, student placement at JAB will enhance the chances of students being placed into courses they prefer as part of their career choice. This is likely to motivate the students, making them work harder and leading to improved performance and improved completion rate. The ANN application may also reduce the cost spend on the application processing and the time the applicants have to wait for the outcome. The ANN application could further increase the chances of high quality applicants getting admission to career courses for which they qualify.

No abstract

This paper presents a design of a system for industry role selection, representing both its structure and behavior. Knowing the right industry role that suits a graduate based on their competences on graduation has remained a critical matter for graduates when searching for jobs after graduation. Thousands of university students graduate each year and enter the market to search for jobs that are limited. Searching without prior information on the most appropriate industry role one is suitable for leads to blind search. Blind search not only puts graduates at risk of long-term unemployment and job mismatch but also overloads employers with many applications during job selection. Therefore, this paper addresses 2 objectives: 1) to model the system’s structure, and 2) to design the algorithm for the system’s behavior. Since object-oriented programming is currently the dominant programming paradigm, object modeling technique was selected to model both the system’s structure and the algorithm for the system’s behavior. To realize object modeling and represent the system’s artifacts in a highly simplified form, Unified Modeling Language (UML) was adopted as the standard modeling toolkit. More specifically, UML class diagram was used to represent the structural model of the system where the underlying objects of the model were exactly similar to those of the problem domain. Finally, use case diagram of the UML toolkit was used to represent the system’s behavior in selecting industry role for graduates. To ensure that the system improves performance of its behavior through experience in selecting industry roles for graduates, Machine Learning (ML) algorithm was designed. Two machine learning techniques, naïve Bayes and Support Vector Machines (SVM), were used as the algorithm’s criteria for selecting industry roles for graduates. Experiments to evaluate performance of the system were conducted using data collected from Software Engineering industry domain. The end product was design of an intelligent industry role selection system with relevant structure and behavior to easily work with both in the academia and industry. Findings reveal the system improves performance of its behavior in selecting industry roles for graduates much better under SVM (67%) than naïve Bayes (57%). On the same benchmark dataset, the system recorded better performance (85%) than reported performance (82%) in the benchmark system. These findings will benefit industry by getting evaluation tool for revealing graduate’s suitability for employment which they can use as prior information for decision making when filtering candidates for interview. Besides, this will provide researchers with a digital platform to study and bridge the gap between industry and academia. Lastly, this will attempt to reduce both low job satisfaction and long-term unemployment that is one of the causes of social and economic pain both in Kenya and around the world. This paper has revealed competence based industry role selection system with relevant structure and behavior can improve searching of jobs by providing a fairly accurate prior information. However, this paper recommends testing this approach with other alternative machine learning techniques as well as other alternative industry domains.

Social media has become part and parcel of students’ life in colleges as it occupies most of their free times. Over time, results of programming units have shown a downward trend among college students at Technical University of Mombasa in Kenya. Programming, like Mathematics, requires a lot of practice that is now consumed by social media. This study investigated the impact of Social Media on Grades in terms of hours spent and the time of day Social Media is used. Descriptive statistical research was used to gain understanding of the predictive power Social Media has on the Programming Grades. A total of 142 students pursuing Degree and Diploma courses in Information Technology participated in the study. The students were drawn from Technical University of Mombasa and Kenya Coast National Polytechnic. Over 90% of the students were active Social Media users. The research was conducted in 2018-2019 academic year. Results show that social media use could predict students who scored high grades in both Degree and Diploma courses. Such students controlled their usage of Social Media. They did not use Social Media in the early morning hours up to afternoons when they were engaged in serious studies. These students used Social Media anytime of the day - which means, in between serious study sessions. On the other hand, more average students used Social Media even in the early morning and up to afternoon sessions. Notably, more average Diploma students use Social Media at bedtime than anytime. Further research is recommended with more data using Machine Learning techniques to develop a model that will predict success or failure in Programming units depending on how one uses social media.

This paper presents a machine learning architecture of a hierarchical model for mapping skills to industry roles. Currently, researchers have been approaching the problem of selecting industry roles for potential employees using flat and top-down methods. Practically, top-down approach is not reliable because it negates the natural mobility of employees in the occupational industry role hierarchy while flat approach does not take advantage of not only the easier learning property of hierarchical approach but also the local information of parent child relationship for better results. The machine learning architecture has been an attempt to address this gap using experimental research design. The mapping model consists of a collection of objects that are hierarchically arranged to progressively group industry role constructs before applying bottom-up approach to select the best. The mapping begins by first selecting the most promising sub-objects at the lower levels before passing this information to the higher levels of the hierarchy to select the most promising functional (main competence), proficiency and specialty (specific competence) objects and eventually the respective constructs.  The end product is an effective machine learning architecture of a model for mapping graduates’ skills to industry roles with relevant attributes to easily work with in the academia and that correctly reflects the hierarchy of industry roles. Findings reveal while SVM (67%) optimizes the model’s accuracy better than naïve Bayes (57%), on the same benchmark dataset the model recorded better performance (85%) than reported performance (82%) in the benchmark model. The findings will benefit industry by getting evaluation tool for revealing information on graduate’s suitability for employment which they can use for decision making when filtering candidates for interview. Besides, this will provide researchers better understanding of the gap between the academia and industry and can use this information to plan on how to bridge the gap using the mapping model. Lastly, this will attempt to reduce both low job satisfaction and long-term unemployment that is one of the causes of social and economic pain both in Kenya and around the world. However, this paper recommends testing this approach with other alternative machine learning techniques as well as other alternative industry domains.