Article Preview
Top1. Introduction
Quality has become a vital subject of discussion among various higher education institutions, and thus it is being extensively studied in recent years (Silva & Fernandes 2011). Today we need to align college activities towards student success, set high expectations, and develop a plan that includes Key Performance Indicators (KPIs). There are five KPIs to measure performance consistently across the college system (2015-16) namely, Graduate Satisfaction, Student Satisfaction, Employer Satisfaction, Graduate Employment, and Graduation Rate (2015-16). This research is oriented towards Graduate Employment and Employer Satisfaction. Mostly the graduate employment or employer satisfaction is determined by marks obtained by students in examinations. The performance is determined by examination achievements assessed with activities such as semester exams, viva, and assignments, etc., (Lakshmi et al. 2013). This has created an environment where formulae, rote learning, marks, and theory are preferred over concepts, critical analysis, knowledge, and practice. Most of the institutions consider marks of the student as the cornerstone of admission to various undergraduate and postgraduate courses and the industry considers final grade obtained by the student during the placement. Thus, this continuum of education ignores other factors like sports, research, and training. The author’s research is centered on the hypothesis that the performance should not be measured only with the marks of the student. The evaluation of students solely on marks has diminished the need to focus on concepts, critical analysis, and practice. Therefore, several other parameters should be identified for evaluating the performance of the students. The other factors that can be considered are projects completed, internships, and skillsets.
The era of Information Technology has led to the collection of enormous volumes of information. Also, the collected data needs a proper technique of taking out the knowledge from large repositories for a better decision-making process. Data mining keeps a close watch on the methodologies for the extraction of useful knowledge from the data and there are many useful data mining tools to extract the knowledge (Yadav et al. 2012). Each university/higher institution has a huge collection of student data like marks, attendance, achievements, etc., the data available with any university are highly unstructured and unlabeled. Therefore, it is necessary to have an insight into the data and observe if there are some patterns present in the available data. The various data mining techniques like clustering, association, prediction, and classification are utilized to understand and find the patterns which are of interest from data. The clustering techniques are widely used to analyze student performance data. In any university or organization, there is no crisp boundary between good student/employee and bad student/employee. There exists some overlap in every organization while categorizing students. For instance, if we consider that a student securing 90% or above is good, and another student who secures 89.9% is equally good and cannot be deemed as Low performer. A soft clustering technique is needed in such situations where there is no clear boundary. Therefore, to address the problem efficiently, the fuzzy c-mean clustering technique is utilized to gain some homogeneous groups of the available heterogeneous population.
This paper aims to understand the comprehensive view of the students’ performance (Sihaloho et al. 2020) that can be obtained by grouping them based on these parameters and concurrently, ascertain the details of their timely performance. The study provides an effective data mining technique for achieving the students’ total performance during the term and help them as well as their teachers to focus on the strategies of improvement by way of monitoring and controlling their performance. The work is further extended by mapping with industry parameters considered during placement activity like functional and industry knowledge, advanced skillset, etc. The rest of the paper is arranged as follows: Section 2 discusses the Literature Review, Section 3 presents the Methodology, Section 4 gives the Experimental Setup, Section 5 discusses the results, and Section 6 discusses the Analysis, Section 7 Mapping the student performance to industry parameters and the conclusion.