Pengembangan Sistem Berbasis Model NER Untuk Verifikasi Dokumen Surat Keputusan Kenaikan Pangkat ASN

Holik, Nur Muhamad (2025) Pengembangan Sistem Berbasis Model NER Untuk Verifikasi Dokumen Surat Keputusan Kenaikan Pangkat ASN. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6022231057-Master_Thesis.pdf Restricted to Repository staff only Download (26MB) | Request a copy

Abstract

Badan Kepegawaian Negara (BKN) bertanggung jawab dalam pembinaan dan penyelenggaraan Manajemen ASN secara nasional, termasuk validasi data pegawai ASN. Salah satu data penting yang harus divalidasi adalah Surat Keputusan (SK) Kenaikan Pangkat. Proses validasi manual yang dilakukan oleh pegawai BKN membutuhkan waktu dan sumber daya yang signifikan, terutama ketika jumlah data usulan yang diterima besar. Penelitian ini mengembangkan model Named Entity Recognition (NER) untuk mengotomatisasi identifikasi entitas dalam SK Kenaikan Pangkat. Model NER yang diusulkan mampu mengenali berbagai entitas penting secara akurat, dengan hasil evaluasi menunjukkan F1-score rata-rata sebesar 99,20\% pada dataset pelatihan dan 98,79\% pada dataset pengujian. Prototipe sistem yang dikembangkan menunjukkan potensi dalam meningkatkan efisiensi validasi SK Kenaikan Pangkat dengan kemampuan ekstraksi informasi yang lebih cepat dan akurat. Hasil ini menunjukkan bahwa penerapan model NER dapat membantu mempercepat dan meningkatkan ketepatan validasi SK Kenaikan Pangkat. ==================================================================================================================================
The State Personnel Agency (BKN) is responsible for fostering and implementing ASN Management nationally, including validating ASN employee data. One of the important data that must be validated is the promotion decision letter (SK). The manual validation process carried out by BKN employees requires significant time and resources, especially when the amount of proposed data received is large.This research develops a Named Entity Recognition (NER) model to automate the identification of entities in the promotion decree. The proposed NER model is able to recognise various important entities accurately, with evaluation results showing an average F1-score of 99.20\% on the training dataset and 98.79\% on the testing dataset. The developed system prototype shows potential in improving the efficiency of validation of promotion decrees with faster and more accurate information extraction capabilities. These results show that the application of the NER model can help accelerate and improve the accuracy of the validation of the promotion decree.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Named-Entity Recognition, Surat Keputusan Kenaikan Pangkat ASN, Validasi Data, ASN promotion decision letter, Data Validation
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.62 Decision support systems T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7882.P3 Pattern recognition systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	NUR MUHAMAD HOLIK
Date Deposited:	09 May 2025 08:54
Last Modified:	09 May 2025 08:54
URI:	http://repository.its.ac.id/id/eprint/119062

Actions (login required)

View Item