Pembentukan Rules untuk Prediksi Kewirausahaan Mahasiswa berdasarkan Theory of Planned Behavior

Nabila, Ifta Jihan (2021) Pembentukan Rules untuk Prediksi Kewirausahaan Mahasiswa berdasarkan Theory of Planned Behavior. Project Report. [s.n.]. (Unpublished)

Text 05111740000034-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Potensi kewirausahaan mahasiswa yang diidentifikasi secara akurat dan dikelola secara optimal akan memberikan kontribusi yang cukup signifikan pada pembangunan sosial dan ekonomi suatu bangsa. Populasi mahasiswa berwirausaha yang sangat sedikit mengakibatkan kondisi ketidakseimbangan kelas data yang berdampak pada ketidakakuratan kinerja machine learning. Pada penelitian ini mengusulkan sebuah teknik preprocessing menggunakan kombinasi teknik sampling SMOTE-NC+IHT dengan teknik seleksi atribut Information Gain untuk mengoptimalkan pemanfaatan database perguruan tinggi nasional sehingga menghasilkan dataset seimbang yang layak untuk teknik machine learning. Berdasarkan evaluasi hasil eksperimen, sebuah model yang mengintegrasikan teknik sampling SMOTE-NC+IHT, metode clustering K-Means dengan algoritma pembentukan rule J48 dapat membangkitkan ruleset berdasarkan Theory of Planned Behavior (TPB) dengan kinerja prediksi yang efektif.
=================================================================================================
The entrepreneurial potential of students who have identified accurately and managed optimally will significantly contribute to the social and economic development of a nation. The very small population of entrepreneurial students results in an imbalance in the class of data, which impacts the inaccuracy of machine learning performance. This study proposes a preprocessing technique using a combination of the SMOTE-NC + IHT sampling technique with the Information Gain attribute selection technique to optimize the national university database's utilization to produce a balanced dataset suitable for machine learning techniques. Based on the evaluation of experimental results, a model that integrates the SMOTE-NC + IHT sampling technique, the K-Means clustering method with the J48 rule formation algorithm can generate ruleset based on the Theory of Planned Behavior (TPB) with effective predictive performance.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	rule generation, entrepreneurial prediction, pembentukan rule, prediksi kewirausahaan
Subjects:	Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) T Technology > T Technology (General) > T57.5 Data Processing
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	IFTA JIHAN NABILA
Date Deposited:	11 Jan 2021 04:42
Last Modified:	17 Jan 2022 08:39
URI:	http://repository.its.ac.id/id/eprint/82405

Actions (login required)

View Item