OCR implementation in archiving system at Borobudur Subdistrict using regular expression and TextRank methods

Abstract

Borobudur Subdistrict is part of a government agency that is required to manage archives, including the registration of incoming letters. Over the past five years, approximately 10,000 incoming letters have been stored in the Borobudur Subdistrict, which must be registered and digitized to support implementing an Electronic-Based Government System. The efforts for registration and digitization have been carried out using conventional methods, which require a considerable amount of time. Therefore, this study aims to develop a system that can assist in the process of archive registration and digitization using Optical Character Recognition (OCR) techniques along with Regular Expression and TextRank methods. The system is designed to extract text from physical documents into digital text through OCR, automatically detecting required text patterns using the Regular Expression method and summarizing documents using the TextRank method. The study's results show a significant increase in efficiency, up to 300%, where the number of archives registered and digitized in one day increased from 20 to 80. This solution proves that implementing this system can significantly improve the speed of the process of archive management in the Borobudur Subdistrict and also provide an effective solution for the registration and digitization process.