Student-Performulator: Predicting Students’ Academic Performance at Secondary and Intermediate Level Using Machine Learning

Annals of Data Science https://doi.org/10.1007/s40745-021-00341-0 Student‑Performulator: Predicting Students’ Academic Performance at Secondary and Intermediate Level Using Machine Learning Shah Hussain1 · Muhammad Qasim Khan1 Received: 22 June 2020 / Revised: 12 April 2021 / Accepted: 21 April 2021 © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2021 Abstract Forecasting academic performance of student has been a substantial research inquest in the Educational Data-Mining that utilizes Machine-learning (ML) procedures to probe the data of educational setups. Quantifying student academic performance is challenging because academic performance of students hinges on several factors. The in hand research work focuses on students’ grade and marks prediction utilizing supervised ML approaches. The data-set utilized in this research work has been obtained from the Board of Intermediate & Secondary Education (B.I.S.E) Peshawar, Khyber Pakhtunkhwa. There are 7 areas in BISEP i.e., Peshawar, FR-Peshawar, Charsadda, Khyber, Mohmand and Upper and Lower Chitral. This paper aims to examine the quality of education that is closely related to the aims of sustainability. The system has created an abundance of data which needs to be properly analyzed so that most useful information should be obtained for planning and future development. Grade and marks forecasting of students with their historical educational record is a renowned and valuable application in the EDM. It becomes an incredible information source that could be utilized in various ways to enhance the standard of education nationwide. Relevant research study reveals that numerous methods for academic performance forecasting are built to carryout improvements in administrative and teaching staff of academic organizations. In the put forwarded approach, the acquired data-set is pre-processed to purify the data quality, the labeled academic historical data of student (30 optimum attributes) is utilized to train regression model and DT-classifier. The regression will forecast marks, while grade will be forecasted by classification system, eventually analyzed the results obtained by the models. The results obtained show that machine learning technology is efficient and relevant for predicting students performance. Keywords Supervised learning · Educational data-mining (EDM) · Machinelearning (ML) · Students’ performance prediction · Learning analytics Extended author information available on the last page of the article 13 Vol.:(0123456789) Annals of Data Science 1 Introduction Quality Education plays an essential role in the Sustainable-Development Goals (17-SDGs), endorsed by United Nations [1]. A crucial aspect to remember while working on sustainable development goals, to provide equal opportunities and sharing it equally. Students’ dissertation in attaining higher education is a serious matter, which need to be evaluated globally. The drop-out students’ ratio from academic institutions causes a loss/resource wastage, which is significant and costly in the educational settings, and affects the evaluation and assessment processes of the academic organizations. The declines in the engineering programs are higher than in all other science and art disciplines, as the study shows [2]. In our aimed study, we will gain the Data Science goals to gain insights from any sort of data. As the Data Science involves developing techniques of storing, recording and analyzing data to extract useful information efficiently. Forecasting analysis of secondary and intermediate students is performed, the marks and grades prediction framework will promote and enhance the standard of education and concentrate on the particular area wherever students do not earn satisfying grades. Education should be given priority in improving our organizations. Administrative and teaching staff should improve their productivity by recognizing curriculum development and skills, which will give students an improved opportunity for learning [3]. In this context, secondary and intermediate-level institutions must also focus on developing and enhancing education models by integrating information and communication technologies (ICT), which can act as an instrument for fostering social accountability and equal opportunities. With this view, the results of ICT in education systems are important as they can make important contributions to the process of learning and teaching as well as to encouraging knowledge building [4]. The use and implementation of ICT application involved in teaching/learning is also known as enhanced learning relying on technology. The term enhanced learning, based on Technology defines the usage of digital technologies to boost learning experience. The utilization of emerging technologies has enhanced learning, it can help to improve critical thinking among students [5]. Technology-based enhancement in learning involves several leading-edge technologies comprises smartphone learning applications, learning-management frameworks, cloud-based learning tools, web-based applications of social networking and social media for education, visual-aids, machinelearning(ML) and DM etc., [6]. In accordance with the impacts of learning-teaching on the sustainability of intermediate & secondary education and tech-boosted learning [5], We must carefully describe the basic information technology requirements that will serve us instead of being a hitch in learning and teaching. For instance, the preparation of teaching and managerial personnel’s for the production of predictive analytic skills as it is crucial for measuring the latent outcomes of the computer-aided framework usage [7]. In to discussed technologies earlier, that is executed with a greater impression in the academic set-ups to produce a large volume of data and save it in means that it could be efficiently presented ubiquitously [8]. The data 13 Annals of Data Science size can exceed the processing volume sometimes, storing and evaluating it with conventional methods. New technologies should be considered in order to perform data analysis such as data mining, intelligent systems, association rules mining, optimization based data mining [9] and big data. The bunch of these novel technologies will enable simple and effective analysis of educational-data, and can be utilized to transform the educational-data in a new shape which could be more beneficial [10–12]. With deep learning mining of educational data is a growing field for research that enable us to analyze and process the educational information collected from different roots [13]. For analyzing educational data, several statistical methods, datamining, visualization and ML gears are utilized. The study analytics generated from academic-data intends to investigate obtained data from the institutional databases. Learning-management frameworks interprets the information, improve learni (...truncated)