Digital Revolution: Using the Naive Bayes Algorithm for Automatic Classification of Archives in a Public Institution in Bandung City

Abstract

The application of the Naive Bayes algorithm in an automatic digital archive classification system at a public institution in Bandung City is covered in this study.  The institution's biggest problem is the growing volume of records, which makes it challenging to manage and find archives rapidly.  To automatically categorize digital archives into predefined categories, a system was developed.  The Naive Bayes algorithm, which provides excellent accuracy, simplicity, and efficiency in processing text data, is the technique employed.  Digital documents that have undergone text-preprocessing steps, including tokenization, stopword removal, and stemming, prior to categorization make up the archive data.  This system uses MySQL, PHP, and the Laravel framework.  According to test results, the system can satisfactorily classify archives while facilitating user location by category.  As a result, this system can improve the institution's digital archive management effectiveness and efficiency.