Step 1: Change a numerical variable into a categorical variable.
Binary (adds a new column):
`newcolumnname = ifelse(columnname <= threshold, "No", "Yes")`
`newdataset = data.frame(dataset, newcolumnname)`

Multi-category (changes original column):
`data$column[criterion] <- 'category name1'`

Step 2: Build a model using the `tree()` function:
`model <- tree(formula = outputcolumn ~ ., data = dataframe)`

Step 3: Predict on test data:
`tree.pred = predict(model, testdata, type = "class")`
`table(tree.pred, testlabel)`

Step 4: Prune the tree using `cv.tree()` to determine optimal complexity:
`cv.carseats = cv.tree(tree.carseats, FUN = prune.misclass)`
The argument `FUN = prune.misclass` uses the classification error rate to guide cross-validation and pruning, rather than the default deviance.

University of Michigan - Ann Arbor

Google

A classification tree applies the decision tree method to classification problems, predicting qualitative (categorical) rather than quantitative responses. It uses recursive binary splitting similar to a regression tree. However, instead of using Residual Sum of Squares (RSS), binary splits are evaluated using criteria like the Gini index or entropy, which measure node purity. The prediction for a given region is typically the most common class among the training observations in that terminal node.

Classification trees

Unlike regression trees, RSS cannot be used as a criterion for making binary splits in classification trees since the most commonly classes, rather than mean values, are used as the predictions. Three criteria that can be used for making binary splits in classification trees are

- classification error rate
- Gini index
- entropy

Criteria for making binary splits in a classification tree

https://towardsdatascience.com/decision-tree-classification-de64fc4d5aac

Learn Before

Related