BORUTA ALGORITHM IS SIGNIFICANT FOR LARGE FEATURE SELECTION OF STUDENT MARKS DATA OF POKHARA UNIVERSITY NEPAL

Author Name: Yagyanath Rimal

Volume: 01 &  Issue:

Country: NEPAL

DOI NO.: NA, DOI Link: http://doi-ds.org/doilink/08.2020-25662434/

Affiliation:

1. School of Engineering, Pokhara University, Nepal
   E-mail:rimal.yagya@gmail.com

ABSTRACT

Boruta algorithm is the best tool for large research data reduction highly used dimension reduction. Although, there was a larger research gap between the data collection and its interoperation, analysis of the appropriate variable of the big database for research generalization. There were many researchers have unknowingly misled their research due to the large set of data before applying the calculation of model execution. Therefore this primary study tries to reduce its features before model deployment, which ultimately significant model up for large data reduction for time and resources execution of machine learning model design. Therefore, this primary reduced 40 independent variables has been reduced largely to 21 variables of Pokhara university student grade data of student grade prediction. So that machine learning model executes quicker, and converse quickly rather than using unimportant variables of large datasets. The machine learning model produces a similar output with 0.7895 accuracies of 95% confidence interval, the p-value is 0.000236 has a significant output of reducing using the boruda algorithm of machine learning design for SGPA prediction of student grades.

Key words: Boruta Algorithm, Principal Component Analysis, Semester Grade Point Average, school leaving certificate