Analisis Implementasi Algoritma Learning Vector Quantization 3 Terhadap Klasifikasi Penentuan Penerima Bantuan Langsung Tunai Desa

The Direct Cash Assistance (BLT) program in Indonesia aims to reduce poverty and improve welfare. However, the recipient selection process often faces social bias, undermining objectivity. This study develops a classification system using the Learning Vector Quantization (LVQ) 3.0 algorithm to support a fairer selection process. The data includes 100 potential recipients from Kelapapati Village, divided into 80% for training and 20% for testing. Parameters include age, education level, home ownership, electricity usage, and water source. Evaluation was conducted using K-Fold Cross Validation to assess algorithm accuracy. The system was tested with various Learning Rate values (0.01; 0.025; 0.05; 0.075; and 0.1) and Window values (0.2; 0.3; 0.4; and 0.5). Results indicate that the LVQ 3.0 algorithm effectively classifies data with 86,22% accuracy, making it suitable for BLT recipient selection. Adding more diverse data is recommended to enhance the system's pattern recognition capability.