Analysis of Water Quality Parameters on the Survival Rate of Vannamei Shrimp using the Random Forest Method

Prih Diantono Abda'u
Ratih Hafsarah Maharrani
Zaenurrohman
Adrian Putra Ramadhan

Jurnal E-Komtek
Vol 8 No 2 (2024)

This study developed a predictive model based on Random Forest algorithm to predict survival rate of Vannamei shrimp using five water quality parameters: dissolved oxygen (DO), temperature, pH, salinity, and Total Dissolved Solids (TDS). The model was trained on this data and evaluated using Mean Squared Error (MSE) and R² Score, with an MSE of 0.71 and R² Score of 1.00. Endpoint testing was conducted using Postman to verify the model response, with output parameters including anomaly_detected, recommendation, and survival rate. The model successfully detected anomalous conditions and provided recommendations according to the detected water quality parameters. Test results showed that DO and salinity had the greatest influence on survival rate, while pH, TDS, and temperature made moderate contributions.