Laporan Kerja Praktek 8 September 2025 - 9 Januari 2026 di Laboratorium Komputasi Berbasis Jaringan (ITS Surabaya)

Mahadhika, Muhammad Izzul Sinar (2026) Laporan Kerja Praktek 8 September 2025 - 9 Januari 2026 di Laboratorium Komputasi Berbasis Jaringan (ITS Surabaya). Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5025221093-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Generasi data sintetis merupakan solusi untuk mengatasi keterbatasan data dalam pengembangan sistem ketika data asli tidak dapat digunakan karena alasan privasi atau regulasi. Penelitian ini mengembangkan sistem generasi data sintetis menggunakan model klasifikasi BERT berbasis CPU yang dapat diakses tanpa memerlukan infrastruktur GPU. Sistem dikembangkan dengan pendekatan: two-stage classification. Model diimplementasikan menggunakan BERT Base Uncased dan dilatih pada dataset nama kolom database. Hasil pengujian berfokus pada akurasi antara prediksi tipe data dan prediksi semantik data dalam klasifikasi data sintetik.
===================================================================================================================================
Synthetic data generation is a solution to overcome data limitations in system development when raw data cannot be used for privacy or regulatory reasons. This research develops a synthetic data generation system using a CPU-based BERT classification model that can be accessed without requiring GPU infrastructure. The system was developed using a two-stage classification approach. The model was implemented using BERT Base Uncased and trained on a database column name dataset. Test results focus on the accuracy between data type prediction and data semantic prediction in synthetic data classification.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	Generasi Data Sintetis, BERT, Klasifikasi Berbasis CPU, Two-Stage Classification, Synthetic Data Generation, BERT, CPU-Based Classification, Two-Stage Classification
Subjects:	Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Information Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	MUHAMMAD IZZUL SINAR MAHADHIKA
Date Deposited:	19 Jan 2026 02:04
Last Modified:	19 Jan 2026 02:04
URI:	http://repository.its.ac.id/id/eprint/129673

Actions (login required)

View Item