Klasifikasi Teks Laporan Masyarakat Pada Situs Lapor! Menggunakan Recurrent Neural Network

Abstract

The existence of public dissatisfaction with public services causes the public to be provided with facilities to make complaints. One of the sites that can be used to make complaints is the Public Service Complaint Management System (SP4N LAPOR!). With this site, complaints made by the public can be handled quickly, transparently and accountably in accordance with the authority of each organizer. However, the large number of complaints that had to be processed caused the process of data verification and sorting of reports by respective departments to take quite a long time, so the report classification process was needed to speed up the handling and follow-up of a report. The purpose of this research is to classify each complaint report from the public in preparation for the verification process of each public report document, which is expected to have an impact on the accelerated process of handling and follow-up of each related institution or agency. In this study, Long Short-Term Memory Recurrent Neural Network was used to perform the classification process for each public report document. The learning model is evaluated using k-fold cross-validation of 10 parts of data. The evaluation results show that the average f-measure percentage is 85.69% for the balanced dataset and 79.44% for the unbalanced dataset, while the highest evaluation value of all evaluations results in an f-measure of 88.82%. The high accuracy of the modeling indicates that the proposed method can be used to classify public report documents.

Evaluasi Model Machine Learning Klasifikasi Gerak Tangan Untuk Sistem Kontrol Prototipe Prostesis Tangan

“…Klasifikasi dan rekognisi pola (classification and pattern recognition) pada serangkaian data (data series) merupakan salah satu pendekatan yang dimuat oleh neural network [10], [11]. Penerapan recurrent neural network (RNN) dapat dijadikan pendekatan pada permasalahan data series, menimbang kemampuan aristektur tidak hanya melihat setiap data point sebagai bagian yang terpisah, namun juga meninjau hubungan antar data point-nya [12]. Merujuk dari hal tersebut, kompleksitas respon tubuh dalam memberikan gerakan dapat meng-implementasikan konsep ini.…”

Implementasi Algoritma Text Mining dan Cosine Similarity untuk Desain Sistem Aspirasi Publik Berbasis Mobile

“…Adapun perbedaannya dengan penelitian lain, Cosine Similarity banyak digunakan dalam mencari persamaan tetapi tidak dengan menggabungkan Text Mining. Kajian terkait membahas aspirasi masyarakat, namun fokusnya hanya pada sistem pelaporan yang menggunakan metode Long Term Memory [13]. Kemudian penelitian yang mengukur dan mengidentifikasi kesamaan dokumen laporan yang menggunakan skema term Class-Based Indexing dengan Cosine similarity [14].…”

Literature Review Klasifikasi Data Menggunakan Metode Cosine Similarity dan Artificial Neural Network

“…Nilai evaluasi rata-rata tertinggi adalah sebesar 88,82%. Dari hasil akurasi tertinggi dapat dilihat bahwa klasifikasi teks dokumen laporan masyarakat dapat menggunakan metode Long Short Term Memory Recurent Neural Network [28].…”