Implementasi Algoritma Adaptive Neurofuzzy Inference System (Anfis) Dalam Memprediksi Komitmen Mahasiswa Baru Ittp

Violita, Anggraini (2022) Implementasi Algoritma Adaptive Neurofuzzy Inference System (Anfis) Dalam Memprediksi Komitmen Mahasiswa Baru Ittp. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover.pdf Download (260kB)
	Text Abstract.pdf Download (81kB)
	Text Abstrak.pdf Download (82kB)
	Text BAB I.pdf Download (81kB)
	Text BAB II.pdf Download (359kB)
	Text BAB III.pdf Download (178kB)
	Text BAB IV.pdf Restricted to Registered users only Download (650kB) | Request a copy
	Text BAB V.pdf Download (88kB)
	Text Daftar Pustaka.pdf Download (211kB)
	Text Lampiran.pdf Restricted to Registered users only Download (642kB) | Request a copy

Abstract

The development of industry in the field of education especially in admission, creates competition between them. Admission in Indonesia have increased from year to year, currently the number of admissions has reached 4482 units. One of the strategies that must be taken by the management at a university is to manage the number of students who enroll in the college. The problem in this research is admission are difficult to predict whether prospective students have a big commitment. This study aims to provide work efficiency and a reliable system as a predictive model in determining student commitment to continue their registration or not, using a prediction method called the Adaptive Neuro Fuzzy Inference System (ANFIS) algorithm. Test results, the value of average error and the average RMSE is getting smaller when the number of iterations increases from 60 to 80 which is worth 37% with value rules 216 and the target error value is 0.1. This method has the ability to predict by reading and learning from input data, where the predictions in this research produce sensitivity values of 80.01%, specificity 70.95% and accuracy 76%. Keywords: Adaptive Neuro fuzzy inference system, Prediction, Student commitment

Actions (login required)

View Item