Klasifikasi Masa Tunggu Alumni Untuk Mendapatkan Pekerjaan Berdasarkan Kompetensi Menggunakan Algoritma C4.5 (Studi Kasus : Fasilkom Unsika)

Rezkika, Febbyola; Sari, Betha Nurina; Irawan, Agung Susilo Yuda

doi:10.35889/progresif.v17i2.652

Cited by 6 publications

(7 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The accuracy obtained in this study is 80.37% and 79.56% of precision. This study also shows that some features have a more significant influence on classification waiting time, such as the city of work, class year, graduation year, GPA, organizational history, expertise certificate, and type of company they work for [5].…”

Section: Introductionmentioning

confidence: 61%

Classification Analysis of Waiting Period for Telkom University Alumni to Get Jobs Using Decision Tree and Support Vector Machine

Miranda¹,

Lhaksamana²

2022

bits

View full text Add to dashboard Cite

Tracer analysis is one of the ways to increase a university's accreditation. Tracer studies, also known as graduate surveys, are beneficial for enhancing learning and developing university curricula. The period it takes graduates to secure employment is a measure of their quality. The sooner graduates obtain a job, the higher their perceived quality. Conversely, if it takes graduates longer to find employment, their quality is deemed lower. To gain new knowledge from the tracer study dataset regarding the relationship between university contribution and alumni capability in the job market, in this study, data mining techniques are used to determine what factors influence the length of time it takes college graduates to find employment. This classification model contains a total of 2288 data instances from the categorical type of dataset. The features are selected using chi-square. Two classification algorithms, Decision Tree and Support Vector Machine, are compared for the best model. This study also used hyperparameter tuning to improve accuracy. The results show decision tree produces higher accuracy compared to the support vector machine. The accuracy obtained from the decision tree model is 55.02% and increased to 65.06% after hyperparameter tuning. Meanwhile, the support vector machine brought an accuracy of 60.40% and increased to 62.15% after hyperparameter tuning. Factors that affect the classification of the alumni waiting period in getting a job in this study are sex, faculty of the study field, department of the study field, study period, company specification, company category, and work location.

show abstract

Section: Introductionmentioning

confidence: 61%

Classification Analysis of Waiting Period for Telkom University Alumni to Get Jobs Using Decision Tree and Support Vector Machine

Miranda¹,

Lhaksamana²

2022

bits

View full text Add to dashboard Cite

show abstract

“…Evaluasi dihasilkan nilai akurasi sebesar 48,629% dengan parameter pengujian 10 Folds Cross-Validation. Penelitian lainnya oleh [9] mengenai prediksi waktu tunggu alumni mendapatkan pekerjaannya dengan menggunakan algoritma decision tree (C4.5) tunggal dan dibandingkan algoritma decision tree (C4.5) dengan fitur Forward Selection. Hasil menunjukkan bahwa algoritma decision tree (C4.5) dengan fitur forward selection meraih performa terbaik dengan nilai Accuracy 80,37%, Precision 79,56%, Recall 81,34%, F1-Score 80,40% dan AUC 0.914 yang termasuk ke dalam kategori excellent Classification.…”

Section: Pendahuluanunclassified

Penerapan Algoritma C4.5 Dengan Feature Forward Selection Untuk Analisis Capaian Indikator Kinerja Utama Berdasarkan Tracer Study

Rahmayani,

Nurina Sari,

Maulana

et al. 2024

jati

View full text Add to dashboard Cite

Perguruan tinggi, sebagai pusat ilmu, penelitian, dan pengabdian masyarakat, dituntut untuk fokus pada capaian Indikator Kinerja Utama Perguruan Tinggi Negeri (IKU-PTN). Salah satu kunci dalam mengelola kinerja perguruan tinggi adalah melalui IKU-PTN yang mendukung relevansi dengan industri dan dunia kerja terdapat pada IKU1. Namun, data tracer study Fasilkom Unsika menunjukkan capaian IKU1 pada tahun 2020 dan 2021 di bawah target, menandakan perlunya analisis pola data alumni. Penelitian ini menggunakan algoritma C4.5 untuk mengevaluasi kelayakan alumni dalam mencapai IKU1. Setelah feature selection, ditemukan atribut paling berpengaruh seperti "Status Bekerja," "Gaji," "Mendapat Pekerjaan Kurang dari 6 Bulan," "1.2 x UMP," dan "Jenis Tempat Bekerja." Model terbaik, yaitu model 5, menghasilkan akurasi 98.77% dengan Weighted Precision rata-rata 100%, Recall 96%, dan f1-Score 89%. Penerapan Decision Tree C4.5 dan metodologi Knowledge Discovery in Databases (KDD) menghasilkan 5 aturan penting yang membantu menilai pencapaian IKU1 berdasarkan fitur-fitur signifikan. Penelitian ini memberikan panduan bagi lembaga pendidikan untuk memahami faktor-faktor yang memengaruhi kesuksesan penempatan kerja alumni serta memberikan prediksi akurat untuk meningkatkan capaian IKU1. Kontribusi ini berpotensi mendorong evaluasi dan pengembangan program pendidikan yang lebih efektif.

show abstract

“…Penelitian terkait dengan prediksi kesiapan kerja mahasiswa dilakukan dengan merujuk pada penelitian prediksi waktu tunggu alumni dalam mendapatkan pekerjaan, Pada penelitian ini dilakukan kombinasi teknik yaitu Supervised Learning dan Unsupervised Learning. Penelitian dengan teknik Supervised Learning yang dilakukan oleh [6,7] dengan menerapkan algoritma C4.5 untuk memprediksi masa tunggu kerja alumni. Penelitian [8] menggunakan Naïve Bayes dengan Backward Elimintaion dan terbukti dapat meningkatkan akurasi, [9,10] menggunakan ANN, [11] menggunakan logistic regression.…”

Section: Pendahuluanunclassified

“…Hasil dari penelitian ini adalah tempat tinggal memiliki pengaruh terhadap IPK menggunakan teknik Uji Parsial. Penelitian [6] melakukan klasifikasi masa tunggu alumni menggunakan algoritma C4.5 dengan fitur forward selection dengan nilai akurasi sebesar 80,37%. Pada penelitian ini, bertujuan untuk melihat kemampuan fitur forward selection yang ternyata lebih baik dabandingkan dengan C4.5 tunggal.…”

Section: Pendahuluanunclassified

Prediksi Kesiapan Kerja Mahasiswa menggunakan Algoritme K-Means dan C4.5

Noviyanto

Mukti

2022

JOSTECH

View full text Add to dashboard Cite

Student work readiness is quite influential on the existence of a university. The waiting time required by students to get a job can affect the mentality of students and the value of higher education from the assumptions of society. This will greatly affect the interest of parents or prospective students to continue their education at universities that have a bad image. Therefore, predictions of student work readiness before graduation are needed for consideration by higher education institutions to overcome the problem of waiting time for student work after graduation. The source of this research data is obtained from the Surakarta University database by utilizing alumni data from tracer studies as train data and 6-semester active student data as test data. The initial step taken is preprocessing to eliminate noise that can interfere with or affect the final result. The research method that will be used is to implement the K-Means and C4.5 algorithms for grouping and prediction processes. The data train used is 150 data and the testing data is 59 data. The results obtained by the K-Means algorithm can cluster 143 data correctly by comparing with the original data. The best cluster value obtained is K = 3.

show abstract

Klasifikasi Masa Tunggu Alumni Untuk Mendapatkan Pekerjaan Berdasarkan Kompetensi Menggunakan Algoritma C4.5 (Studi Kasus : Fasilkom Unsika)

Cited by 6 publications

References 0 publications

Classification Analysis of Waiting Period for Telkom University Alumni to Get Jobs Using Decision Tree and Support Vector Machine

Classification Analysis of Waiting Period for Telkom University Alumni to Get Jobs Using Decision Tree and Support Vector Machine

Penerapan Algoritma C4.5 Dengan Feature Forward Selection Untuk Analisis Capaian Indikator Kinerja Utama Berdasarkan Tracer Study

Prediksi Kesiapan Kerja Mahasiswa menggunakan Algoritme K-Means dan C4.5

Contact Info

Product

Resources

About