<![CDATA[Failure in catfish farming is often caused by not finding the best composition combination when you want to start cultivation such as, types of ponds, antiseptic administration, etc. Therefore we need a system that aims to find the best combination of parameters while predicting the condition of the fish before it is implemented in the real world. One method that can be applied to predict fish conditions is certainty factor. However, the performance of certainty factor is highly dependent on experts related to the problem so that the resulting solution is vulnerable to being trapped in the local optimum area. One approach that can be used to overcome this problem is to apply optimization algorithms, namely Particle Swarm Optimization (PSO). PSO explores the search space to find the value of the expert cf based on the value of the particle cost. The value of the cost is designed to minimize the distance between random values ​​and the weight value so that the smaller is close to 0 (zero) the greater the chance of a particle being selected as a solution. This study uses hybrid Particle Swarm Optimization-Certainty Factor to identify the condition of catfish. The quality of Certainty Factor is evaluated using test data from experts by comparing system output. The experimental results show that the PSO-Certainty Factor hybrid algorithm produces better predictive results compared to the Certainty Factor algorithm which is 90%.]]>
Copyrights © 2019
