The manual filing of letters at SMPK Ratu Damai makes the storage, retrieval, and grouping of letters ineffective and carries the risk of misclassification. This study aims to apply the Decision Tree algorithm to a digital archiving system to automatically classify letters based on archive categories. The research methods include data collection, text preprocessing consisting of case folding, tokenizing, stopword removal, and stemming, TF-IDF weighting, and the development of a Decision Tree model. The dataset used consisted of 228 letter documents. The system was developed as a web-based system using PHP, MySQL, XAMPP, and Visual Studio Code. It was then evaluated using a Confusion Matrix with Accuracy, Precision, Recall, and F1-Score metrics. The implementation results showed that the system was able to automate the letter classification process, accelerate document searches from 300 seconds to 8 seconds (97.33% efficiency), and support more structured archive management. Based on the evaluation results, the Decision Tree model demonstrated strong performance with 84.78% Accuracy, 85.79% Precision, 86.09% Recall, and 85.94% F1-Score. ABSTRAK Pengarsipan surat yang masih dilakukan secara manual di SMPK Ratu Damai menyebabkan proses penyimpanan, pencarian, dan pengelompokan surat menjadi kurang efektif serta berisiko menimbulkan kesalahan klasifikasi. Penelitian ini bertujuan menerapkan algoritma Decision Tree pada sistem pengarsipan digital untuk mengklasifikasikan surat secara otomatis berdasarkan kategori arsip. Metode penelitian meliputi pengumpulan data, preprocessing teks yang terdiri atas case folding, tokenizing, stopword removal, dan stemming, pembobotan Term Frequency-Inverse Document Frequency (TF-IDF), serta pembentukan model Decision Tree. Dataset yang digunakan berjumlah 228 dokumen surat periode 2018–2023. Sistem dikembangkan berbasis web menggunakan PHP, MySQL, XAMPP, dan Visual Studio Code, kemudian dievaluasi menggunakan Confusion Matrix dengan metrik Accuracy, Precision, Recall, dan F1-Score. Hasil implementasi menunjukkan bahwa sistem mampu mengotomatisasi proses klasifikasi surat, mempercepat pencarian dokumen dari 300 detik menjadi 8 detik (efisiensi 97,33%), serta mendukung pengelolaan arsip yang lebih terstruktur. Berdasarkan hasil evaluasi, model Decision Tree menunjukkan kinerja yang sangat baik dengan Akurasi 84,78%, Presisi 85,79%, Recall 86,09%, dan F1-Score 85,94%, sehingga layak diterapkan sebagai solusi pengarsipan digital di SMPK Ratu Damai.
Copyrights © 2026