我在R语言中尝试在一个小数据集上拟合决策树模型，但无论输入什么数据集，它总是预测类别标签为Yes。

数据

outlook <- c("sunny", "sunny", "overcast", "rain", "rain", "rain", "overcast", "sunny", "sunny", "rain", "sunny", "overcast", "overcast", "rain")temperature <- c("hot", "hot", "hot", "mild", "cool", "cool", "cool", "mild", "cool", "mild", "mild", "mild", "hot", "mild")humidity <- c("high", "high", "high", "high", "normal", "normal", "normal", "high", "normal", "normal", "normal", "high", "normal", "high")wind <- c("weak", "strong", "weak", "weak", "weak", "strong", "strong", "weak", "weak", "weak", "strong", "strong", "weak", "strong")class <- c("no", "no", "yes", "yes", "yes", "no", "yes", "no", "yes", "yes", "yes", "yes", "yes", "no")data <- data.frame(outlook, temperature, humidity, wind, class)data

编码数据

outlook_new <- as.numeric(as.factor(outlook))temperature_new <- as.numeric(as.factor(temperature))humidity_new <- as.numeric(as.factor(humidity))wind_new <- as.numeric(as.factor(wind))class_new <- as.numeric(as.factor(class))data_new <- data.frame(outlook_new, temperature_new, humidity_new, wind_new, class_new)data_new

构建模型

model <- rpart(class_new ~ ., data=data_new)

创建测试数据点

test_data <- data.frame(outlook_new = 2, temperature_new = 2, humidity_new = 1, wind_new = 1)test_data

预测

predict(model, test_data, type='response')

预测函数总是给出结果为yes，无论输入是什么。

问题出在哪里？

回答：

对于如此小的训练集，你需要更新模型控制，并且对结果持保留态度！

model <- rpart(class ~ ., data = data, control = rpart.control(minsplit = 1))predict(model, newdata = data, type = 'class')#   1   2   3   4   5   6   7   8   9  10  11  12  13  14 #  no  no yes yes yes  no yes  no yes yes yes yes yes  no # Levels: no yes