Predicting the Efficiency and Success Rate of Programming Courses in MOOC Using Machine Learning Approach for Future Employment in the IT Industry

Predicting the Efficiency and Success Rate of Programming Courses in MOOC Using Machine Learning Approach for Future Employment in the IT Industry

Shivangi Gupta, A. Sai Sabitha, Sunil Kumar Chowdhary
Copyright: © 2021 |Pages: 18
DOI: 10.4018/JITR.2021040102
(Individual Articles)
No Current Special Offers


Modern businesses and jobs in demand have witnessed the requirement of programming skills in candidates, for example, business analyst, database administrator, software engineer, software developer, and many more. Programming courses are a very influential and important part of forming the future of the IT industry. Throughout the recent years, a substantial amount of research has been conducted to improve the programming novices, but the problems are returning in every new generation and reporting high failure rates. The dataset used in this study is the ‘CodeChef competition' dataset and the ‘Coursera' dataset. Firstly, this research work conducts the preview analysis to understand the performance of learners in programming languages. Secondly, this work proposes a clear rationale between the popularity of MOOC courses and low completion rates. There is increasingly high enrolment in MOOC courses but with non-ideal completion rates. Finally, it builds the machine learning model and validates the accuracy of the trained model.
Article Preview

1. Introduction

In today’s digital advanced era, every business organization is associated with different technologies, working cultures, and a combination of processes (Gupta et al. 2019). Recent years have observed new businesses and jobs that require individuals to maintain programming language knowledge (Konecki 2014). Information technology jobs that demand coding skills include business analyst, data analyst, data scientist, software developer, software engineer, full-stack developer, database admin, system administrator, and network engineers. The demand in these mentioned careers has been predicted to increase by 12 percent from 2018 to 2026. These professions are predicted to create 546,200 jobs in the near future (US. Department of Labor 2018). By virtue, of the significance of programming languages in the IT sector, information technology departments need learners to complete programming courses that are associated with them to achieve future career requirements efficiently. For IT sector success and to prepare students or learners for the working world, it is essential to find the needs of learners in programming courses and providing adequate support to them. Massive open online courses have outreached millions of learners across the globe. Business schools, engineering universities, and IT companies have also introduced their students and employees towards massive open online courses (MOOC). Anytime, Anywhere and Anybody features have attracted millions of learners towards MOOC courses (Zhang et al. 2019).

Despite, the need for programming languages in IT jobs, learners’ who have little or no experience in programming incur difficulty in learning programming languages and find it hard to complete the programming courses (Bashir and Hoque 2016). This difficulty results in dropout from programming courses of MOOC platforms. (Gupta and Sabitha 2019) have identified the significant attributes (number of events in which learner participated, the total number of days the learner was active, course videos played or not, and total chapters of course material a learner explored) that can help in predicting early dropout from MOOC courses. Several studies have found the factors affecting the dropout rate in MOOC courses (Bharara 2018; Arora 2017; Sabitha 2016) and dropout factors in programming courses (Cigdem and Yildirim 2014; Dang et al. 2016; Wiggins et al. 2017) respectively but, still, there is inadequacy to predict the success rate of courses on MOOC platform. These MOOC programs might be seen as free courses for learners’, but for those MOOC development platforms or institutions (udemy, coursera, udacity, Edureka, Simplilearn, and nptel) who design the course structure for online learning involves high expenditure. The cost may diverge from US$50000 – US$100000 (Bates 2013). Therefore, to tackle this challenge, this research work builds the model to predict the success rate of programming courses on the MOOC platform which will help the IT sector in gaining future talent needs for upcoming jobs.

Figure 1.

Business areas of programming


Besides, to retain the success rate of programming courses in the MOOC environment, it is important to ensure that the correct amount of programming skills must be made compulsory to teach at the school level to influence student’s perception towards learning programming skills. Non-professionals need to learn basic programming skills to have better insights and understanding of project management, policy-making understanding, and business administration (Refer to Figure 1). So, the purpose of this study is also to find out the prerequisite essential for learning programming and coding to be well inclined to future profession needs. Understanding the importance of programming skills could help the IT sector and students with efficient and effective acquisition talent and obtaining jobs in future demanding professions.

The objective of this substantial research work done is as follows:

  • (1)

    Conduct the preview analysis to identify the performance of students in different programming languages in coding competitions / MOOC courses.

  • (2)

    To identify the significant factors responsible for student dropout in programming courses from massive open online courses platform to train the machine learning model.

  • (3)

    Training and building model using a machine learning approach in predicting the efficiency of programming courses on MOOC platform to increase the future career prerequisites required by the IT sector and reduce the loss acquired by the MOOC provider industries.

Complete Article List

Search this Journal:
Volume 16: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 15: 6 Issues (2022): 1 Released, 5 Forthcoming
Volume 14: 4 Issues (2021)
Volume 13: 4 Issues (2020)
Volume 12: 4 Issues (2019)
Volume 11: 4 Issues (2018)
Volume 10: 4 Issues (2017)
Volume 9: 4 Issues (2016)
Volume 8: 4 Issues (2015)
Volume 7: 4 Issues (2014)
Volume 6: 4 Issues (2013)
Volume 5: 4 Issues (2012)
Volume 4: 4 Issues (2011)
Volume 3: 4 Issues (2010)
Volume 2: 4 Issues (2009)
Volume 1: 4 Issues (2008)
View Complete Journal Contents Listing