Heart failure clinical Data Analysis
Download 428.37 Kb.
|
466 project
Support Vector MachineSupport vector machine (SVM) is a type of supervised machine learning algorithm, that analyze the data for classification and regression analysis. Each data item is shown as a point in n-dimensional space in this algorithm, with the value of each function being the value of each coordinate. Then, classification is done by determining the strongest differentiating hyperplane for the two classes. This model is used by most state-of-the-art approaches [8][9][10] for classification purpose. For experiment, we have 0.7 for training of the model and 0.3 for testing of the model. Our experiment shows that we have achieved accuracy of 0.83. The confusion matrix of the SVM is shown in Table 1. # Importing the dataset df = read.csv('heart_failure_clinical_records_dataset.csv',header = TRUE) install.packages("caret") library('caret') intrain <- createDataPartition(y = df$DEATH_EVENT, p= 0.7, list = FALSE) training <- df[intrain,] testing <- df[-intrain,] training[["DEATH_EVENT"]] = factor(training[["DEATH_EVENT"]]) trctrl <- trainControl(method = "repeatedcv", number = 10, repeats = 3) svm_Linear <- train(DEATH_EVENT ~., data = training, method = "svmLinear", trControl=trctrl, preProcess = c("center", "scale"), tuneLength = 10) svm_Linear test_pred <- predict(svm_Linear, newdata = testing) test_pred confusionMatrix(table(test_pred, testing$DEATH_EVENT))
Table 1: Confusion Matrix Download 428.37 Kb. Share with your friends:
|