Time Series Prediction on College Graduation Using KNN Algorithm

Abstract

The KNN algorithm is an algorithm for classifying data based on learning data taken from k of its closest neighbors. Classification using the K-Nearest Neighbor (KNN) algorithm can be used to predict whether a student will graduate on time or even be at risk of dropping out. This research implemented KNN algorithm because of its effectiveness in training large and robustness on noisy data. The input used is in the form of student academic data and produces output, namely the accuracy of the KNN algorithm. Data will be divided into time series into four parts, namely 1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">st</sup> year, 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nd</sup> year, 3 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">rd</sup> year, and 4th year. The time series prediction aims to find out the exact time to make predictions. Testing was conducted using K-Fold Cross Validation by dividing the set of data into several folds, one-fold as test data and the other fold as training data. The results of this test are the accuracy of the predictions of each year experiencing increase and prediction in time series can be used for early detection.