Abstract: Students can choose institutions or universities that fit their academic and career aspirations with the aid of a higher education recommendation system that uses the KNN algorithm. A machine learning technique called the KNN algorithm finds the K closest neighbours to a given data point based on a similarity metric. The algorithm in a recommendation system uses this similarity measure to identify the K institutions that are the most similar to a student's choices and suggests them as prospective possibilities. The KNN algorithm needs a dataset of colleges with information on their location, tuition costs, admission rate, student- faculty ratio, and programmes they offer in order to develop a recommendation system for higher education.The student's preferences must also be reflected in terms of these characteristics. The KNN method can then be used to discover the K universities that are closest to the student's preferences and offer them as potential options. The KNN method frequently uses the Euclidean distance as the similarity metric. We can determine the K universities that are closest to a student's choices by measuring the distance between each institution in the dataset and the student's preferences. These K universities may then be suggested to the student. Cross- validation or holdout validation techniques can be used to assess the performance of the recommendation system. In cross-validation, the dataset is divided into k-folds, the model is trained on k-1 folds, and it is then tested on the final fold. Holdout validation involves training the model on the training set, then testing it on the testing set after randomly partitioning the dataset into training and testing sets. In conclusion, a KNN-based recommendation system for higher education can help students choose colleges that will best suit their academic and professional objectives.
Keywords: Higher education, recommendation system, K-nearest neighbours (KNN) algorithm,machine learning.
|
DOI:
10.17148/IJARCCE.2023.124143
