The practicality of digital document processing has made various companies and organizations switch physical documents to digital. However, the process of extracting data from physical documents manually requires effort and is prone to input errors due to human error. Optical Character Recognition (OCR) technology can be a solution to this problem. OCR is used to recognize letters or characters in an image, and then stored into text data on a computer. In this research, the implementation of OCR technology on a web-based application with Long Short-Term Memory method. Based on accuracy testing, the average error value at the character level is 6.56% and at the word level is 9.98%. From the results obtained, it shows that the application of OCR technology with Long Short-Term Memory method on web applications can be the right solution in the process of extracting data from physical document.
Copyrights © 2023