Implementation Internet of Things for Feeding Catfish Water Quality Analysis Using Linear Regression and K-Nearest Neighbor

Abstract

Cultivation of catfish (Clarias Gariepinus) is a promising business field and also a very productive activity because public interest in catfish is high. This factor is observed by market demand for catfish which is increasing from year to year. In catfish farming, you must pay attention to the acidity of the water (pH), temperature, and oxygen levels, which can change if too much feed is given. This can cause catfish seedlings to die and affect the catfish harvest. Catfish farmers often provide excessive food which causes many catfish seeds to die. This research will conduct a study on an Internet Of Things technology that can be used to monitor the acidity level in water pH, temperature, and oxygen levels as well as feed fish. The Internet of Things is very influential for monitoring the quality of catfish ponds by distributing information data resulting from sensor monitoring. The data obtained will be predicted for water quality in the pond by implementing a Linear Regression method. Furthermore, the acquisition of data from the predictions that have been carried out will be processed again to go to the next phase, namely classifying with the K-Nearest Neighbor algorithm method to carry out the identification phase of water types based on the nearest neighbors. This prediction is used to anticipate and notify catfish farmers through applications if there is a water acidity level (pH), temperature, and oxygen and feed levels that have run out