<![CDATA[Water is a chemical compound that is very important for every living thing for survival on this earth. On earth water has a very large area compared to the mainland this compound has an area of ​​almost 71% around the land. The water also consists of sea water, rivers, lakes, ground water, swamp water, snow and steam which are in the air layer that contains mineral substances that are recruited in water. To determine the classification of water quality status using the Support Vector Machine and Naive Bayes methods. This method was chosen because previous studies get high accuracy results for classification. The parameters used are the degree of acidity (pH), TDS, NO2, NO3, hardness, chloride, manganese. The Vector Machine and Naive Bayes Suppord method will provide the results of the comparison of the accuracy of the two methods. Testing on the system is done using the K-Fold Cross Valadation test with the highest accuracy results when K = 9 for the Support Vector Machine method and K = 5 for the Naive Bayes method. Testing parameters for the Support Vector Machine method gets the highest accuracy when the threshold value is , C = 3, γ = 0.01, λ = 2.5, maximum iteration value = 1, σ = 0.1. From these tests the accuracy of the Support Vector Machine method was 78.70% and the Naive Bayes method was 85.78%. The best results obtained by the classification of water quality status are the Naive Bayes method compared to using the Support Vector Machine method because the average accuracy of the Naive Bayes method is higher. Water is a chemical compound that is very important for every living thing for survival on this earth. On earth water has a very large area compared to the mainland this compound has an area of ​​almost 71% around the land. The water also consists of sea water, rivers, lakes, ground water, swamp water, snow and steam which are in the air layer that contains mineral substances that are recruited in water. To determine the classification of water quality status using the Support Vector Machine and Naive Bayes methods. This method was chosen because previous studies get high accuracy results for classification. The parameters used are the degree of acidity (pH), TDS, NO2, NO3, hardness, chloride, manganese. The Vector Machine and Naive Bayes Suppord method will provide the results of the comparison of the accuracy of the two methods. Testing on the system is done using the K-Fold Cross Valadation test with the highest accuracy results when K = 9 for the Support Vector Machine method and K = 5 for the Naive Bayes method. Testing parameters for the Support Vector Machine method gets the highest accuracy when the threshold value is , C = 3, γ = 0.01, λ = 2.5, maximum iteration value = 1, σ = 0.1. From these tests the accuracy of the Support Vector Machine method was 78.70% and the Naive Bayes method was 85.78%. The best results obtained by the classification of water quality status are the Naive Bayes method compared to using the Support Vector Machine method because the average accuracy of the Naive Bayes method is higher.]]>
Copyrights © 2019
