Open AccessAn android-based application to predict student with extraordinary academic achievement
Author(s) -
I Gede Iwan Sudipa,
I Nyoman Saputra Wahyu Wijaya,
Made Leo Radhitya,
I M A Mahawan,
I N A Arsana
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1469/1/012043
Subject(s) - graduation (instrument) , mathematics education , accreditation , computer science , dropout (neural networks) , sample (material) , academic year , psychology , medical education , mathematics , machine learning , medicine , chemistry , geometry , chromatography
The study period of students who pass the time limit and high numbers of dropout in a college can affect the value of campus accreditation. The anticipation of that possibility, the college must make predictions about potential students don’t graduate on time. This study aims to build a system capable of predicting students who have the potential. If students with unpredictable graduation risks can be identified in the early stages, then the indication of dropout rates may be reduced by providing special appeals to students at risk. Prediction analysis applies the K-Nearest Neighbors method to dig up the trace data stack and look for the proximity of the data with the new data. The test data used student class of 2011 with 100 students as sample data. This method of classification is based on several attributes, namely the evaluation of the 1st semester to semester 6th, the number of GPA, credits that have been taken each semester, number of credits passed, and the number of credits that didn’t pass. The result of classification becomes the output of the system which is then entered into the testing phase. This stage compares the output with the original data with 70, 73% accuration result.