ORIGINAL RESEARCH Application of hybrid clustering methods for student performance evaluation Ramjeet Singh Yadav 1 Received: 2 April 2018 / Accepted: 23 April 2018 Ó Bharati Vidyapeeth’s Institute of Computer Applications and Management 2018 Abstract In the present research paper, an attempt has been made to design for evaluating the assessment of stu- dent in the educational environment by using the hybrid clustering methods. Various issues along problems are associated with the academic performance evaluation in the field of education. The proposed new approach known as hybrid clustering is based on integrated techniques of Subtractive and Fuzzy C-Means clustering methods. The assessment of student academic performance can be con- sidered as a clustering problem. The clusters are formed on the basis of the intelligence level of students. The intelli- gence level wise grouping is essential for maintaining the homogeneity of the group. Otherwise it would be difficult to provide good educational services to the highly diverse student population. Keywords Fuzzy logic  Subtractive Clustering  Fuzzy C-Means  Academic performance  Fuzzy inference 1 Introduction Homogenous objects having same features can be grouped by the application of data clustering technique utilizing similarity index consider in objects group. Data cluster analysis involves the study of group structures with no prior knowledge and thus falls into the category of unsupervised learning. The underlying data abstraction can exposed using clustering algorithm, which can be further utilized for partitioning the input space of fuzzy systems. The fuzzy clustering method allows objects to several clusters simultaneously with different membership degrees [1]. The major concern associated with clustering is to discover the organization patterns into groups to reveal similarities and differences to draw conclusions [2, 3]. An Expert System (ES) belongs to the highly specialized domain having encoded database prepared by human expert which can be deployed in solving problems in a very narrow window of problem domain [4]. ES utilizes a basic set of rules in the form IF-THEN or other suitable construct to build rules which match certain assumptions from knowledge data bases provided by human experts for rea- soning and referencing. The knowledge database is pre- pared by aggregating numerous streams of data published or digitally. ES mimics natural language processing (NLP) because it utilizes symbolic representation to represent knowledge and inference through applying similar sym- bolic pattern (rules, networks and frames). K-Means clus- tering method is an iterative algorithm which involves partitioning and clustering data in a crisp pattern [5]. A point can be assigned with partial membership to multiple clusters by using Fuzzy C-Means (FCM) algorithm which is generalization of classical C-Means (hard C-Means) algorithm [6]. It produces a generalized hard partition (soft partition) of a known set of data. By applying proper constrains it produces a constrained generalized hard par- tition (constrained soft partition) [7, 8]. Regression consists in exploring the association between dependent and inde- pendent variables to identify their impact on dependent variable [9]. Bayesian classifiers assume feature value independence on a given class with respect to other & Ramjeet Singh Yadav ramjeetsinghy@gmail.com 1 Department of Computer Science and Engineering, Ashoka Institute of Technology and Management, Engineering Chauraha, Paharia, Sarnath, Varanasi, Uttar Pradesh 221007, India 123 Int. j. inf. tecnol. https://doi.org/10.1007/s41870-018-0192-2