<![CDATA[The productivity of Etawa goats is influenced by various factors, including physical condition, environment, and care provided. This study proposes a Decision Tree (DT) algorithm optimized using Particle Swarm Optimization (PSO) to more accurately predict Etawa goat productivity. A case study was conducted at the "Bang Usman" farm in Pekalongan Regency, Central Java, Indonesia, using secondary data collected from January to December 2015. The dataset comprises 500 samples with 13 attributes, including goat age, body weight, body height, housing condition, feeding accuracy, feed type, reproductive cycle, daily milk production, ear type, horn shape, coat color, forehead shape, and mating calendar. Predicting productivity is a critical data-mining technique for measuring success in agricultural industries. The standard Decision Tree achieved an accuracy of 92.60%. By integrating PSO for attribute selection and parameter optimization, the PSO-DT model demonstrates measurable improvements: accuracy increased by 1.20%, precision by 3.04%, and recall by 0.68%. The model also identifies the eight most influential attributes to support farm decision-making. This research contributes a validated PSO-DT model that outperforms standard DT, offering practical value for livestock management and serving as a reference for future studies on optimization algorithms in agricultural productivity prediction in Indonesia.]]>
Copyrights © 2026
