Applying Neuro Fuzzy System to Analyze Durian Minerals within Soil for Precision Agriculture

The objective of this paper is to study the analysis of precision farming for farmers who grow durian in Thailand. By using data from the environment within the plot which consists of composition of minerals within the soil pH, temperature, humidity and amount of sunlight. The data analyzed from the past 5 years consisted of 1,826 records. For data analysis, the researchers proposed the Neuro Fuzzy system algorithm to improve the results during for recommendations and then compared with 4 algorithms, neural network, decision tree, K-Nearest Neighbor and Naive Bayes. The results found that neural network had a precision of 89.05%. The decision tree has an accuracy of 84.12%. K-Nearest Neighbor has an accuracy of 83.03% and Naive Bayes has an accuracy of 87.23%. The proposed algorithm application provides an accuracy of 91.24%.