Informatics Publishing Limited

A. R. Mohamed Shanavas ¹, M. S. Mythili ², D. Regina Margaret ²

Affiliations
1 Department of Computer Science, Jamal Mohamed College, Trichy, IN
2 Department of Computer Applications, Bishop Heber College, Trichy, IN

Abstract

Knowledge and skill of the stake holders are the most dominant factors to evaluate their quality of education. The aim of the paper is to analyze the performance of students of selected UG departments who have consistently figured in the top ten ranks in the first five semester marks by correlating the college marks with their higher secondary marks. The study also includes the correlation of male and female, rural and urban, residential and non residential students. The analysis is done by Statistical Data Mining, Regression Techniques and Hypothesis testing. Data mining is the key to the Knowledge Discovery Database (KDD).. The results of this analysis help the educational institutions to admit quality students for further studies and to retain the good results with high quality education.

Keywords

Data Mining, KDD, DSS, Hypothesis Testing, Prediction.

Full Text

     

M. S. Mythili ¹, A. R. Mohamed Shanavas ²

Affiliations
1 Bharathidasan University and Department of Computer Applications, Bishop Heber College, Tiruchirappalli - 620017, Tamil Nadu, IN
2 Department of Computer Science, Jamal Mohamed College, Tiruchirappalli - 620020, Tamil Nadu, IN

Abstract

Background: The main objective of this paper is to focus on the early prediction of occurrence of cognitive disorders such as Autism, Dyslexia and Delirium among children. The common primary attributes are related to learning, social interaction, behaviour, understanding of objects and so on. Detecting this disorder at an early age is challenge lying among health care specialists, and researchers. Methodology: The proposed prediction method involves the modelling approach such as Meta Heuristic and Fuzzy Cognitive Map named as MEHECOM. Findings: The primary aim of MEHECOM model is to identify the disorders among children as an early measure and support towards early mechanism to alleviate from these disorders. The performance shows that MEHECOM predicts chances of dyslexia and autism disorders at an average of 65.22% compared to 48.7% of FEAST which adopts fuzzy cognitive map and 35.92% of decision tree approach. Applications: This MEHECOM model can be applied for the earlier prediction of all other cognitive disorders like amnesia, dementia etc.

Keywords

Cognitive Disorders, Early Prediction, Fuzzy Cognitive Map, Meta Heuristic Approach

Full Text