Finalizing Naive Bayes

.~lock.accidentsFull.csv#

@@ -1 +0,0 @@
-,noah,NovaArchSys,29.03.2023 17:55,file:///home/noah/.config/libreoffice/4;

Schrick-Noah_Learning-Practice-8.ipynb

@@ -578,11 +578,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 202,
    "metadata": {},
    "outputs": [],
    "source": [
-    "# c."
+    "# c.\n",
+    "trainData, validData = train_test_split(accidents_df, test_size=0.4, random_state=26)"
    ]
   },
   {
@@ -595,12 +596,23 @@
    ]
   },
   {
-   "cell_type": "code",
+   "attachments": {},
-   "execution_count": null,
+   "cell_type": "markdown",
    "metadata": {},
-   "outputs": [],
    "source": [
-    "# i"
+    "# i. \n",
+    "HOUR_I_R \n",
+    "ALIGN_I \n",
+    "WRK_ZONE \n",
+    "WKDY_I_R \n",
+    "INT_HWY \n",
+    "LGTCON_I_R \n",
+    "PROFIL_I_R \n",
+    "SPD_LIM \n",
+    "SUR_CON \n",
+    "TRAF_CON_R \n",
+    "TRAF_WAY \n",
+    "WEATHER_R"
    ]
   },
   {
@@ -609,16 +621,53 @@
    "metadata": {},
    "source": [
     "# ii. \n",
-    "Run a naive Bayes classifier on the complete training set with the relevant predictors (and INJURY as the response). Note that all predictors are categorical. Show the confusion matrix.\n"
+    "Run a naive Bayes classifier on the complete training set with the relevant predictors (and INJURY as the response). Note that all predictors are categorical. Show the confusion matrix."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 209,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Training\n",
+      "Confusion Matrix (Accuracy 0.5291)\n",
+      "\n",
+      "       Prediction\n",
+      "Actual    0    1\n",
+      "     0 4197 8195\n",
+      "     1 3724 9193\n"
+     ]
+    }
+   ],
    "source": [
-    "# ii."
+    "# ii.\n",
+    "predictors = ['HOUR_I_R', 'ALIGN_I', 'WRK_ZONE', 'WKDY_I_R', 'INT_HWY',\n",
+    "               'LGTCON_I_R', 'PROFIL_I_R', 'SPD_LIM', 'SUR_COND', \n",
+    "               'TRAF_CON_R', 'TRAF_WAY', 'WEATHER_R']\n",
+    "\n",
+    "X = pd.get_dummies(accidents_df[predictors])\n",
+    "y = accidents_df['Injury']\n",
+    "\n",
+    "X_train, X_valid, y_train, y_valid = train_test_split(X, y, test_size=0.40, random_state=1)\n",
+    "\n",
+    "# run naive Bayes\n",
+    "delays_nb = MultinomialNB(alpha=0.01)\n",
+    "delays_nb.fit(X_train, y_train)\n",
+    "\n",
+    "# predict probabilities\n",
+    "predProb_train = delays_nb.predict_proba(X_train)\n",
+    "predProb_valid = delays_nb.predict_proba(X_valid)\n",
+    "\n",
+    "# predict class membership\n",
+    "y_valid_pred = delays_nb.predict(X_valid)\n",
+    "y_train_pred = delays_nb.predict(X_train)\n",
+    "\n",
+    "print(\"Training\")\n",
+    "classificationSummary(y_train, y_train_pred) "
    ]
   },
   {
@@ -627,16 +676,32 @@
    "metadata": {},
    "source": [
     "# iii. \n",
-    "What is the overall error for the validation set?\n"
+    "What is the overall error for the validation set?"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 210,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Validation\n",
+      "Confusion Matrix (Accuracy 0.5288)\n",
+      "\n",
+      "       Prediction\n",
+      "Actual    0    1\n",
+      "     0 2838 5491\n",
+      "     1 2460 6085\n"
+     ]
+    }
+   ],
    "source": [
-    "# iii."
+    "# iii.\n",
+    "print(\"Validation\")\n",
+    "classificationSummary(y_valid, y_valid_pred) "
    ]
   },
   {
@@ -650,11 +715,21 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 217,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.057\n"
+     ]
+    }
+   ],
    "source": [
-    "# iv."
+    "# iv.\n",
+    "pctg_inc = round(100* abs(0.5288 - 0.5291)/(0.5291), 3)\n",
+    "print(pctg_inc)\n"
    ]
   },
   {
@@ -667,12 +742,12 @@
    ]
   },
   {
-   "cell_type": "code",
+   "attachments": {},
-   "execution_count": null,
+   "cell_type": "markdown",
    "metadata": {},
-   "outputs": [],
    "source": [
-    "# v."
+    "The probability is rounded to 0 due to the extremely low likelihood of sustaining an injury at such low speeds. \n",
+    "The pivot tables display values ranging from E-6 to E-9, which is assumed as 0."
    ]
   }
  ],