{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tree-Based Methods"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "New R commands\n",
    "- `ifelse(cond, a, b)` take a if cond is satisfied else b.\n",
    "- `tree(formula, dataset` fit a tree to the dataset.\n",
    "- `cv.tree(tree, FUN = )` perform cross-validation on a tree with loss function FUN.\n",
    "- `prune.misclass(tree, best = k)` and `prune.tree(tree, best = k)` prune a tree to have only the best k terminal nodes.\n",
    "\n",
    "Useful to remember\n",
    "- `table(predicted, actual)` creates a confusion matrix (true/false positives/negatives)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fitting Classification Trees"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The tree library is used to construct classification and regression trees."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "library(tree)\n",
    "library(ISLR)\n",
    "attach(Carseats)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We first use classification trees to analyze the Carseats data set. In these\n",
    "data, Sales is a continuous variable, and so we begin by recoding it as a\n",
    "binary variable. We use the ifelse() function to create a variable, called\n",
    "High, which takes on a value of Yes if the Sales variable exceeds 8, and\n",
    "takes on a value of No otherwise.\n",
    "\n",
    "Finally, we use the data.frame() function to merge High with the rest of\n",
    "the Carseats data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "High=ifelse(Sales<=8,\"No\",\"Yes\")\n",
    "Carseats=data.frame(Carseats,High)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now use the tree() function to fit a classification tree in order to predict\n",
    "High using all variables but Sales. The syntax of the tree() function is quite\n",
    "similar to that of the lm() function. The \".-Sales\" on the right-hand-side means\n",
    "that we take all predictors of Carseats except Sales."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree.carseats=tree(High~.-Sales,Carseats)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The summary() function lists the variables that are used as internal nodes\n",
    "in the tree, the number of terminal nodes, and the (training) error rate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary(tree.carseats)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that the training error rate is 9%. For classification trees, the deviance reported in the output of summary() is given by $-2\\sum_m\\sum_kn_{mk}\\log\\hat p_{mk}$ where $n_{mk}$ is the number of observations in the mth terminal node that belong to the kth class.\n",
    "\n",
    "A small deviance indicates a tree that provides a good fit to the (training) data. The residual mean deviance reported is simply the deviance divided by $n−|T_0|$, which in this case is 400−27 = 373.\n",
    "\n",
    "One of the most attractive properties of trees is that they can be\n",
    "graphically displayed. We use the plot() function to display the tree structure, and the text() function to display the node labels. The argument\n",
    "pretty=0 instructs R to include the category names for any qualitative predictors, rather than simply displaying a letter for each category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot(tree.carseats)\n",
    "text(tree.carseats,pretty=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we just type the name of the tree object, R prints output corresponding\n",
    "to each branch of the tree. R displays the split criterion (e.g. Price<92.5 ), the\n",
    "number of observations in that branch, the deviance, the overall prediction\n",
    "for the branch ( Yes or No ), and the fraction of observations in that branch\n",
    "that take on values of Yes and No . Branches that lead to terminal nodes are\n",
    "indicated using asterisks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree.carseats"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to properly evaluate the performance of a classification tree on\n",
    "these data, we must estimate the test error rather than simply computing\n",
    "the training error. We split the observations into a training set and a test\n",
    "set, build the tree using the training set, and evaluate its performance on\n",
    "the test data. The predict() function can be used for this purpose. In the\n",
    "case of a classification tree, the argument type=\"class\" instructs R to return\n",
    "the actual class prediction. This approach leads to correct predictions for\n",
    "around 70.5 % of the locations in the test data set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "set.seed(4)\n",
    "train=sample(1:nrow(Carseats), 200)\n",
    "Carseats.test=Carseats[-train,]\n",
    "High.test=High[-train]\n",
    "tree.carseats=tree(High~.-Sales,Carseats,subset=train)\n",
    "tree.pred=predict(tree.carseats,Carseats.test,type=\"class\")\n",
    "table(tree.pred,High.test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(89+52)/200"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we consider whether pruning the tree might lead to improved\n",
    "results. The function cv.tree() performs cross-validation in order to\n",
    "determine the optimal level of tree complexity; cost complexity pruning\n",
    "is used in order to select a sequence of trees for consideration. We use\n",
    "the argument FUN=prune.misclass in order to indicate that we want the\n",
    "classification error rate to guide the cross-validation and pruning process,\n",
    "rather than the default for the cv.tree() function, which is deviance. The\n",
    "cv.tree() function reports the number of terminal nodes of each tree considered ( size ) as well as the corresponding error rate and the value of the\n",
    "cost-complexity parameter used (k , which corresponds to α in the slides and in section (8.4) of the book)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "set.seed(3)\n",
    "cv.carseats=cv.tree(tree.carseats,FUN=prune.misclass)\n",
    "names(cv.carseats)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cv.carseats"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that, despite the name, dev corresponds to the cross-validation error\n",
    "rate in this instance. The tree with 10 terminal nodes results in the lowest\n",
    "cross-validation error rate, with 50 cross-validation errors. We plot the error\n",
    "rate as a function of both size and k."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "par(mfrow=c(1,2))\n",
    "plot(cv.carseats$size,cv.carseats$dev,type=\"b\")\n",
    "plot(cv.carseats$k,cv.carseats$dev,type=\"b\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now apply the prune.misclass() function in order to prune the tree to\n",
    "obtain the ten-node tree."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "prune.carseats=prune.misclass(tree.carseats,best=10)\n",
    "plot(prune.carseats)\n",
    "text(prune.carseats,pretty=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How well does this pruned tree perform on the test data set? Once again,\n",
    "we apply the predict() function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree.pred=predict(prune.carseats,Carseats.test,type=\"class\")\n",
    "table(tree.pred,High.test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(92+58)/200"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now 75 % of the test observations are correctly classified, so not only has\n",
    "the pruning process produced a more interpretable tree, but it has also\n",
    "improved the classification accuracy.\n",
    "\n",
    "If we increase the value of best, we obtain a larger pruned tree with lower\n",
    "classification accuracy:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "prune.carseats=prune.misclass(tree.carseats,best=16)\n",
    "plot(prune.carseats)\n",
    "text(prune.carseats,pretty=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree.pred=predict(prune.carseats,Carseats.test,type=\"class\")\n",
    "table(tree.pred,High.test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(89+52)/200"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fitting Regression Trees"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we fit a regression tree to the Boston data set. First, we create a\n",
    "training set, and fit the tree to the training data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\n",
       "Regression tree:\n",
       "tree(formula = medv ~ ., data = Boston, subset = train)\n",
       "Variables actually used in tree construction:\n",
       "[1] \"rm\"    \"lstat\" \"crim\"  \"age\"  \n",
       "Number of terminal nodes:  7 \n",
       "Residual mean deviance:  10.38 = 2555 / 246 \n",
       "Distribution of residuals:\n",
       "    Min.  1st Qu.   Median     Mean  3rd Qu.     Max. \n",
       "-10.1800  -1.7770  -0.1775   0.0000   1.9230  16.5800 "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "library(MASS)\n",
    "set.seed(1)\n",
    "train = sample(1:nrow(Boston), nrow(Boston)/2)\n",
    "tree.boston=tree(medv~.,Boston,subset=train)\n",
    "summary(tree.boston)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notice that the output of summary() indicates that only four of the variables have been used in constructing the tree. In the context of a regression\n",
    "tree, the deviance is simply the sum of squared errors for the tree. We now\n",
    "plot the tree."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot(tree.boston)\n",
    "text(tree.boston,pretty=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The variable rm is the average number of rooms per dwelling.\n",
    "The variable lstat measures the percentage of individuals with lower\n",
    "socioeconomic status. The tree indicates that lower values of lstat correspond to more expensive houses. \n",
    "The tree predicts a median house price\n",
    "of 21380 USD for smaller homes in suburbs in which residents have high socioeconomic status ( rm<6.543 and lstat<14.405 )\n",
    "\n",
    "Now we use the cv.tree() function to see whether pruning the tree will\n",
    "improve performance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cv.boston=cv.tree(tree.boston)\n",
    "plot(cv.boston$size,cv.boston$dev,type='b')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this case, the most complex tree is selected by cross-validation. How-\n",
    "ever, if we wish to prune the tree, we could do so as follows, using the\n",
    "prune.tree() function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "prune.boston=prune.tree(tree.boston,best=5)\n",
    "plot(prune.boston)\n",
    "text(prune.boston,pretty=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "35.2868818594623"
      ],
      "text/latex": [
       "35.2868818594623"
      ],
      "text/markdown": [
       "35.2868818594623"
      ],
      "text/plain": [
       "[1] 35.28688"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0gAAANICAIAAAByhViMAAAACXBIWXMAABJ0AAASdAHeZh94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\nAAAAlSDYAQAAqATBDgAAQCX+D7tIpOXsFEauAAAAAElFTkSuQmCC",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 420,
       "width": 420
      },
      "text/plain": {
       "height": 420,
       "width": 420
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "yhat=predict(tree.boston,newdata=Boston[-train,])\n",
    "boston.test=Boston[-train,\"medv\"]\n",
    "plot(yhat,boston.test)\n",
    "abline(0,1)\n",
    "mean((yhat-boston.test)^2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In other words, the test set MSE associated with the regression tree is\n",
    "35.3. The square root of the MSE is therefore around 5.94, indicating\n",
    "that this model leads to test predictions that are within around 5940 USD of\n",
    "the true median home value for the suburb."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "ir"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}