Sistem Prediksi Jumlah Perkara Tindak Pidana Umum Menggunakan Metode Backpropagation Neural Network

Abstract

The preparation of the Ministry/Agency Work Plan and Budget (RKA-K/L) is an annual obligation for every government work unit to ensure accountability and effectiveness of performance. For the District Attorney’s Office of Kendal, an accurate estimation of the volume of general criminal case handling is a key indicator in determining the allocation of operational budgets. However, the high fluctuation and uncertainty in the number of cases each year require a more precise projection method compared to conventional manual calculations. This study aims to implement artificial intelligence technology to predict the target number of cases as an instrument for budget monitoring and evaluation. The method used is an Artificial Neural Network (ANN) with the Backpropagation algorithm. The network architecture applied is based on a 7-7-1 model, integrating 7 neurons in the input layer, 7 neurons in the hidden layer, and 1 neuron in the output layer. The results show that the Backpropagation ANN model is capable of recognizing historical patterns of case handling data effectively, achieving an accuracy level of 75.38%. This value indicates that the developed model is sufficiently reliable to assist the District Attorney’s Office of Kendal in projecting annual workloads. Through this measurable estimation, the distribution of prosecution operational budgets can be carried out more optimally, transparently, and accurately to support fair law enforcement.