Andika, Albert Agung (2024) Implementasi Fitur Pada MyITS StudentConnect Untuk Memverifikasi Dokumen Portofolio Secara Otomatis Menggunakan Optical Character Recognition. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
5027201069-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2026. Download (4MB) | Request a copy |
Abstract
MyITS StudentConnect merupakan salah satu media bagi Institut Teknologi Sepuluh Nopember (ITS) dalam mendata mahasiswa ITS yang mengikuti kegiatan akademik atar nonakademik di luar jadwal perkuliahan. Dalam melakukan pendataan, diperlukan tenaga bagi para dosen wali untuk melakukan konfirmasi atas beragam bentuk portofolio data yang dimasukkan oleh para mahasiswa. Untuk membantu dosen wali dalam melakukan verifikasi, diajukan sebuah fitur verifikasi otomatis pada MyITS StudentConnect menggunakan Optical Character Recognition (OCR). OCR melakukan pengambilan data atas portofolio kompetisi berupa surat tugas dan sertifikat yang memiliki format PDF dan JPG. Model OCR yang digunakan adalah PyTesseract dan EasyOCR. Model OCR dievaluasi menggunakan empat metrik, yaitu accuracy, precision, recall, dan f1 score. Sistem verifikasi dokumen portofolio mahasiswa pada MyITS StudentConnect menggunakan EasyOCR untuk sertifikat dan PyTesseract untuk surat tugas, mencapai akurasi 55%-80% pada sertifikat dan 80%-86% pada surat tugas. Integrasi dengan MyITS StudentConnect melalui API memudahkan proses verifikasi, memungkinkan klasifikasi otomatis dokumen yang memerlukan tinjauan lebih lanjut oleh dosen wali.
=================================================================================================================================
MyITS StudentConnect is a platform used by the Institut Teknologi Sepuluh Nopember (ITS) to track students who participate in academic and non-academic activities outside of regular class schedules. To facilitate the data recording process, faculty advisors need to confirm various types of portfolio data submitted by students. To assist faculty advisors in verification, an automatic verification feature using Optical Character Recognition (OCR) is proposed for MyITS StudentConnect. OCR captures data from competition portfolios such as assignment letters and certificates in PDF and JPG formats. The OCR models used are PyTesseract and EasyOCR. These OCR models are evaluated using four metrics: accuracy, precision, recall, and F1 score. The student portfolio document verification system on MyITS StudentConnect uses EasyOCR for certificates and PyTesseract for assignment letters, achieving an accuracy of 55%-80% for certificates and 80%-86% for assignment letters. Integration with MyITS StudentConnect through an API simplifies the verification process, allowing for automatic classification of documents that require further review by faculty advisors.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | OCR, PyTesseract, EasyOCR, Document Verification, Portfolio, MyITS, StudentConnect |
Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T174 Technological forecasting |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information Technology > 59201-(S1) Undergraduate Thesis |
Depositing User: | Albert Agung Andika |
Date Deposited: | 06 Feb 2024 04:16 |
Last Modified: | 06 Feb 2024 04:16 |
URI: | http://repository.its.ac.id/id/eprint/106213 |
Actions (login required)
View Item |