A Data Mining-Based Approach for Exploiting the Characteristics of University Lecturers
The faculty evaluation forms can be considered as valuable data source to exploit knowledge which helps to improve the quality of teaching and learning in universities. In this paper, we analyze previous studies on exploiting faculty evaluation forms acco
- PDF / 862,034 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 67 Downloads / 149 Views
Abstract. The faculty evaluation forms can be considered as valuable data source to exploit knowledge which helps to improve the quality of teaching and learning in universities. In this paper, we analyze previous studies on exploiting faculty evaluation forms according to major problems and their solutions. On that basis, we propose and solve the problem of mining useful knowledge about human resource of Ton Duc Thang University using a data mining-based approach. The experimental data are collected from the online faculty evaluation system of our university, with more than 140,000 evaluation forms. We apply the solution to analyze the data set and draw meaningful comments for the characteristics of the lecturers so that human resource can be exploited and constructed appropriately and efficiently. The results obtained are compared to a previous study on clustering lecturers based on performance and correlation coefficient analysis method. Keywords: Student feedback Faculty performance evaluation Faculty evaluation form Clustering
Teaching performance
1 Introduction The quality of education has always been considered as the foundation of the long-term development of all countries. In order to provide people with sufficient knowledge and skills to labor market and enhance their reputation, universities must constantly improve the quality of teaching and learning. Many strategies have been applied to measure the faculty performance, including: student ratings, peer ratings, self-evaluation, videos, student interviews, exit and alumni ratings, employer ratings, administrator ratings, teaching scholarships, teaching awards, learning outcome measures and teaching portfolio [2]. Among these strategies, student ratings are considered as the most popular evaluation tool [4]. In this paper, based on the analysis of previous studies on the exploitation of knowledge from faculty evaluation forms to improve the quality of teaching and support stakeholders such as administrators, lecturers and students in making decisions, we propose a new problem that exploits evaluation forms to obtain useful knowledge about human resource of our university and the method to solve that problem. On that basis, administrators can make decisions in salary increase and task assignment; students can choose appropriate lecturers; lecturers realize their strengths and weaknesses. © Springer International Publishing Switzerland 2016 P. Perner (Ed.): ICDM 2016, LNAI 9728, pp. 41–53, 2016. DOI: 10.1007/978-3-319-41561-1_4
42
T.-D. Do et al.
We apply the proposed solution on a real data set including 143,117 forms from the online faculty evaluation system of Ton Duc Thang University. The results obtained are compared to the only study on clustering lecturers based on performance and correlation coefficient analysis method. It provides an overview of human resource in our university, laying the foundation for the exploitation and development of human resource efficiently. The main contributions of our work are the following: • • • •
Construct
Data Loading...
