updated/Predict number of 911 calls in each block group.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Workflow: Predict number of 911 calls in each US block group.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> * 🔬 Data Science\n",
    "* 📈 Statistics\n",
    "* 👟 Ready To Run!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Background**: The 911 has been designated as the \"<font color=green>Universal Emergency Number</font>\" for citizens throughout the United States to request emergency assistance In 1968, 911 became the national emergency number for the United States The calls are placed by a Voice over Internet Protocol (”VoIP”) End-User to Carrier’s Internet Protocol (”IP”) services"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Importance**:\n",
    "\n",
    "- A good understanding of expected 911 calls might help with resource allocation for 911 call centers\n",
    "- Reduce the stress of the 911 operators, who suﬀer from signiﬁcant employee turnover\n",
    "- Help strategy makers anticipate the occurrences of emergencies and enable them to effectively handle the emergency by appropriate allocation of resources"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Technology**:\n",
    "\n",
    "1. Programming Languages: Python\n",
    "2. Data Visualization: Seaborn"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Connect-to-your-ArcGIS-online-organization\" data-toc-modified-id=\"Connect-to-your-ARCGIS-online-organization-1\">Connect to your ArcGIS online organization</a></span></li><li><span><a href=\"#Use-overlay-tools-to-combine-the-911-calls-with-US-block-groups\" data-toc-modified-id=\"Use-overlay-tools-to-combine-the-911-calls-with-US-block-groups-2\">Use overlay tools to combine the 911 calls with US block groups</a></span></li><li><span><a href=\"#Train-and-test-your-model\" data-toc-modified-id=\"Train-and-test-your-model-3\">Train and test your model</a></span></li><li><span><a href=\"#Enriching-the-mont_block_groups-with-demographic-variables\" data-toc-modified-id=\"Enriching-the-mont_block_groups-with-demographic-variables-4\">Enriching the mont_block_groups with demographic variables</a></span></li><li><span><a href=\"#Train-and-test-your-model-to-check-improvement\" data-toc-modified-id=\"Train-and-test-your-model-to-check-improvement-5\">Train and test your model to check improvement</a></span></li><li><span><a href=\"#Lets-add-more-demographic-variables-using-the-enrich_layer-tool.\" data-toc-modified-id=\"Lets-add-more-demographic-variables-using-the-enrich_layer-tool.-6\">Lets add more demographic variables using the enrich_layer tool.</a></span></li><li><span><a href=\"#Results-after-enriching-data\" data-toc-modified-id=\"Results-after-enriching-data-7\">Results after enriching data</a></span></li><li><span><a href=\"#Visualization\" data-toc-modified-id=\"Visualization-8\">Visualization</a></span></li></ul></div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Workflow using ArcGIS API for Python"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Connect to your ArcGIS online organization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from arcgis import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "gis = GIS(\"https://deldev.maps.arcgis.com\", \"demos_deldev\", \"DelDevs12\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's search **911 calls in Montgomery county** layer using the `search()` method from the content property of your gis object.you can pass a parameter **outside_org=True** to access the contents outside your organization."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "montgomery_calls = gis.content.search('title: \"Montgomery 911 calls\"', 'Feature layer', outside_org=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Import **dislay** module to see the list of searched items."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.display import display"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"item_container\" style=\"height: auto; overflow: hidden; border: 1px solid #cfcfcf; border-radius: 2px; background: #f6fafa; line-height: 1.21429em; padding: 10px;\">\n",
       "                    <div class=\"item_left\" style=\"width: 210px; float: left;\">\n",
       "                       <a href='https://deldev.maps.arcgis.com/home/item.html?id=67497a184b8f4bb69109f77d576440f7' target='_blank'>\n",
       "                        <img src='data:image/png;base64,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' width='200' height='133' class=\"itemThumbnail\">\n",
       "                       </a>\n",
       "                    </div>\n",
       "\n",
       "                    <div class=\"item_right\"     style=\"float: none; width: auto; overflow: hidden;\">\n",
       "                        <a href='https://deldev.maps.arcgis.com/home/item.html?id=67497a184b8f4bb69109f77d576440f7' target='_blank'><b>Join Features to Montgomery 911 calls</b>\n",
       "                        </a>\n",
       "                        <br/>Feature layer generated from Join Features<img src='https://deldev.maps.arcgis.com/home/js/jsapi/esri/css/images/item_type_icons/featureshosted16.png' style=\"vertical-align:middle;\">Feature Layer Collection by arcgis_python\n",
       "                        <br/>Last Modified: June 05, 2018\n",
       "                        <br/>0 comments, 9 views\n",
       "                    </div>\n",
       "                </div>\n",
       "                "
      ],
      "text/plain": [
       "<Item title:\"Join Features to Montgomery 911 calls\" type:Feature Layer Collection owner:arcgis_python>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div class=\"item_container\" style=\"height: auto; overflow: hidden; border: 1px solid #cfcfcf; border-radius: 2px; background: #f6fafa; line-height: 1.21429em; padding: 10px;\">\n",
       "                    <div class=\"item_left\" style=\"width: 210px; float: left;\">\n",
       "                       <a href='https://deldev.maps.arcgis.com/home/item.html?id=db3a0520ca3149fca19413b5dd23e9b1' target='_blank'>\n",
       "                        <img src='http://static.arcgis.com/images/desktopapp.png' class=\"itemThumbnail\">\n",
       "                       </a>\n",
       "                    </div>\n",
       "\n",
       "                    <div class=\"item_right\"     style=\"float: none; width: auto; overflow: hidden;\">\n",
       "                        <a href='https://deldev.maps.arcgis.com/home/item.html?id=db3a0520ca3149fca19413b5dd23e9b1' target='_blank'><b>Intersect_of_Montgomery_911_calls_and_USA_Block_Groups</b>\n",
       "                        </a>\n",
       "                        <br/>Feature layer generated from Overlay layers<img src='https://deldev.maps.arcgis.com/home/js/jsapi/esri/css/images/item_type_icons/featureshosted16.png' style=\"vertical-align:middle;\">Feature Layer Collection by demos_deldev\n",
       "                        <br/>Last Modified: December 17, 2018\n",
       "                        <br/>0 comments, 0 views\n",
       "                    </div>\n",
       "                </div>\n",
       "                "
      ],
      "text/plain": [
       "<Item title:\"Intersect_of_Montgomery_911_calls_and_USA_Block_Groups\" type:Feature Layer Collection owner:demos_deldev>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div class=\"item_container\" style=\"height: auto; overflow: hidden; border: 1px solid #cfcfcf; border-radius: 2px; background: #f6fafa; line-height: 1.21429em; padding: 10px;\">\n",
       "                    <div class=\"item_left\" style=\"width: 210px; float: left;\">\n",
       "                       <a href='https://deldev.maps.arcgis.com/home/item.html?id=3b723ebf35e14e4195e5dcf35685678e' target='_blank'>\n",
       "                        <img src='data:image/png;base64,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' width='200' height='133' class=\"itemThumbnail\">\n",
       "                       </a>\n",
       "                    </div>\n",
       "\n",
       "                    <div class=\"item_right\"     style=\"float: none; width: auto; overflow: hidden;\">\n",
       "                        <a href='https://deldev.maps.arcgis.com/home/item.html?id=3b723ebf35e14e4195e5dcf35685678e' target='_blank'><b>Montgomery 911 calls</b>\n",
       "                        </a>\n",
       "                        <br/><img src='https://deldev.maps.arcgis.com/home/js/jsapi/esri/css/images/item_type_icons/featureshosted16.png' style=\"vertical-align:middle;\">Feature Layer Collection by demo_deldev\n",
       "                        <br/>Last Modified: May 30, 2018\n",
       "                        <br/>0 comments, 33 views\n",
       "                    </div>\n",
       "                </div>\n",
       "                "
      ],
      "text/plain": [
       "<Item title:\"Montgomery 911 calls\" type:Feature Layer Collection owner:demo_deldev>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for item in montgomery_calls:\n",
    "    display(item)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get the **montgomery 911 calls** item."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "mont_911_calls_item = montgomery_calls[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"item_container\" style=\"height: auto; overflow: hidden; border: 1px solid #cfcfcf; border-radius: 2px; background: #f6fafa; line-height: 1.21429em; padding: 10px;\">\n",
       "                    <div class=\"item_left\" style=\"width: 210px; float: left;\">\n",
       "                       <a href='https://deldev.maps.arcgis.com/home/item.html?id=3b723ebf35e14e4195e5dcf35685678e' target='_blank'>\n",
       "                        <img src='data:image/png;base64,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' width='200' height='133' class=\"itemThumbnail\">\n",
       "                       </a>\n",
       "                    </div>\n",
       "\n",
       "                    <div class=\"item_right\"     style=\"float: none; width: auto; overflow: hidden;\">\n",
       "                        <a href='https://deldev.maps.arcgis.com/home/item.html?id=3b723ebf35e14e4195e5dcf35685678e' target='_blank'><b>Montgomery 911 calls</b>\n",
       "                        </a>\n",
       "                        <br/><img src='https://deldev.maps.arcgis.com/home/js/jsapi/esri/css/images/item_type_icons/featureshosted16.png' style=\"vertical-align:middle;\">Feature Layer Collection by demo_deldev\n",
       "                        <br/>Last Modified: May 30, 2018\n",
       "                        <br/>0 comments, 33 views\n",
       "                    </div>\n",
       "                </div>\n",
       "                "
      ],
      "text/plain": [
       "<Item title:\"Montgomery 911 calls\" type:Feature Layer Collection owner:demo_deldev>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mont_911_calls_item"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since the item is a Feature Layer Collection, accessing the layers property will give us a list of FeatureLayer objects."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "mont_911_calls = mont_911_calls_item.layers[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since we want to predict number of 911 calss in block group for the year 2017.<br>\n",
    "Our prediction must be based on 2016 data.<br>\n",
    "Apply filter on the layer to keep only 2016 data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "mont_911_calls.filter = \"timeStamp BETWEEN '2016-01-01' AND '2016-09-01'\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we want to predict number of 911 calls in each block group of the montgomery county,<br> let's use the `search` method again to find US block group layer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"item_container\" style=\"height: auto; overflow: hidden; border: 1px solid #cfcfcf; border-radius: 2px; background: #f6fafa; line-height: 1.21429em; padding: 10px;\">\n",
       "                    <div class=\"item_left\" style=\"width: 210px; float: left;\">\n",
       "                       <a href='https://deldev.maps.arcgis.com/home/item.html?id=d3a11165055747068a5a456c2e2f0e31' target='_blank'>\n",
       "                        <img src='data:image/png;base64,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' width='200' height='133' class=\"itemThumbnail\">\n",
       "                       </a>\n",
       "                    </div>\n",
       "\n",
       "                    <div class=\"item_right\"     style=\"float: none; width: auto; overflow: hidden;\">\n",
       "                        <a href='https://deldev.maps.arcgis.com/home/item.html?id=d3a11165055747068a5a456c2e2f0e31' target='_blank'><b>USA Block Groups</b>\n",
       "                        </a>\n",
       "                        <br/>This layer presents the census block groups of the United States in the 50 states, the District of Columbia, and Puerto Rico.<img src='https://deldev.maps.arcgis.com/home/js/jsapi/esri/css/images/item_type_icons/featureshosted16.png' style=\"vertical-align:middle;\">Feature Layer Collection by esri_dm\n",
       "                        <br/>Last Modified: October 31, 2018\n",
       "                        <br/>0 comments, 11,658 views\n",
       "                    </div>\n",
       "                </div>\n",
       "                "
      ],
      "text/plain": [
       "<Item title:\"USA Block Groups\" type:Feature Layer Collection owner:esri_dm>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "block_groups = gis.content.search('title: \"USA Block Groups\"', 'Feature layer', outside_org=True)[0]\n",
    "block_groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "mont_block_groups = block_groups.layers[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Filter the block groups by \"STCOFIPS = 42091\" since FIPS code for montgomery county is 42091."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "mont_block_groups.filter = \"STCOFIPS = 42091\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<FeatureLayer url:\"https://services.arcgis.com/P3ePLMYs2RVChkJx/arcgis/rest/services/USA_Block_Groups/FeatureServer/0\">"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mont_block_groups"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Use overlay tools to combine the 911 calls with US block groups\n",
    "overlay_layers combines two or more layers into one single layer using either intersection, union  or erase method. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from arcgis.features.manage_data import overlay_layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "intersect_blkgrp_with_calls = overlay_layers(mont_911_calls, mont_block_groups, tolerance=0, context={},\n",
    "                           output_name=\"Combine 911 calls with blkgrp\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"item_container\" style=\"height: auto; overflow: hidden; border: 1px solid #cfcfcf; border-radius: 2px; background: #f6fafa; line-height: 1.21429em; padding: 10px;\">\n",
       "                    <div class=\"item_left\" style=\"width: 210px; float: left;\">\n",
       "                       <a href='https://deldev.maps.arcgis.com/home/item.html?id=c8a555fab468454899215691bc8c70fa' target='_blank'>\n",
       "                        <img src='data:image/png;base64,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' width='200' height='133' class=\"itemThumbnail\">\n",
       "                       </a>\n",
       "                    </div>\n",
       "\n",
       "                    <div class=\"item_right\"     style=\"float: none; width: auto; overflow: hidden;\">\n",
       "                        <a href='https://deldev.maps.arcgis.com/home/item.html?id=c8a555fab468454899215691bc8c70fa' target='_blank'><b>Combine 911 calls with blkgrp</b>\n",
       "                        </a>\n",
       "                        <br/><img src='https://deldev.maps.arcgis.com/home/js/jsapi/esri/css/images/item_type_icons/featureshosted16.png' style=\"vertical-align:middle;\">Feature Layer Collection by demo_deldev\n",
       "                        <br/>Last Modified: December 14, 2018\n",
       "                        <br/>0 comments, 3 views\n",
       "                    </div>\n",
       "                </div>\n",
       "                "
      ],
      "text/plain": [
       "<Item title:\"Combine 911 calls with blkgrp\" type:Feature Layer Collection owner:demo_deldev>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intersect_blkgrp_with_calls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "intersect_layer = intersect_blkgrp_with_calls.layers[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<FeatureLayer url:\"https://services6.arcgis.com/SMX5BErCXLM7eDtY/arcgis/rest/services/Combine 911 calls with blkgrp/FeatureServer/0\">"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intersect_layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdf = pd.DataFrame.spatial.from_layer(intersect_layer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['AGE_10_14', 'AGE_15_19', 'AGE_20_24', 'AGE_25_34', 'AGE_35_44',\n",
       "       'AGE_45_54', 'AGE_55_64', 'AGE_5_9', 'AGE_65_74', 'AGE_75_84',\n",
       "       'AGE_85_UP', 'AGE_UNDER5', 'AMERI_ES', 'ASIAN', 'AVE_FAM_SZ',\n",
       "       'AVE_HH_SZ', 'BLACK', 'BLKGRP', 'CNTY_FIPS', 'FAMILIES', 'FEMALES',\n",
       "       'FHH_CHILD', 'FIPS', 'HAWN_PI', 'HISPANIC', 'HOUSEHOLDS', 'HSEHLD_1_F',\n",
       "       'HSEHLD_1_M', 'HSE_UNITS', 'Join_Count', 'MALES', 'MARHH_CHD',\n",
       "       'MARHH_NO_C', 'MED_AGE', 'MED_AGE_F', 'MED_AGE_M', 'MHH_CHILD',\n",
       "       'MULT_RACE', 'OBJECTID_1', 'OTHER', 'OWNER_OCC', 'ObjectID',\n",
       "       'POP10_SQMI', 'POP12_SQMI', 'POP2010', 'POP2012', 'RENTER_OCC', 'SHAPE',\n",
       "       'SQMI', 'STATE_FIPS', 'STCOFIPS', 'TARGET_FID', 'TRACT', 'VACANT',\n",
       "       'WHITE', 'addr', 'desc_', 'e', 'lat', 'lng', 'timeStamp', 'title',\n",
       "       'twp', 'zip'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sdf.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AGE_10_14</th>\n",
       "      <th>AGE_15_19</th>\n",
       "      <th>AGE_20_24</th>\n",
       "      <th>AGE_25_34</th>\n",
       "      <th>AGE_35_44</th>\n",
       "      <th>AGE_45_54</th>\n",
       "      <th>AGE_55_64</th>\n",
       "      <th>AGE_5_9</th>\n",
       "      <th>AGE_65_74</th>\n",
       "      <th>AGE_75_84</th>\n",
       "      <th>...</th>\n",
       "      <th>WHITE</th>\n",
       "      <th>addr</th>\n",
       "      <th>desc_</th>\n",
       "      <th>e</th>\n",
       "      <th>lat</th>\n",
       "      <th>lng</th>\n",
       "      <th>timeStamp</th>\n",
       "      <th>title</th>\n",
       "      <th>twp</th>\n",
       "      <th>zip</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>59</td>\n",
       "      <td>67</td>\n",
       "      <td>71</td>\n",
       "      <td>145</td>\n",
       "      <td>168</td>\n",
       "      <td>157</td>\n",
       "      <td>244</td>\n",
       "      <td>63</td>\n",
       "      <td>232</td>\n",
       "      <td>257</td>\n",
       "      <td>...</td>\n",
       "      <td>1286</td>\n",
       "      <td>CEDAR RD &amp; WARING RD</td>\n",
       "      <td>CEDAR RD &amp; WARING RD;  CHELTENHAM; Station 358...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.071914</td>\n",
       "      <td>-75.118497</td>\n",
       "      <td>2016-01-04 02:22:12</td>\n",
       "      <td>EMS: HEMORRHAGING</td>\n",
       "      <td>CHELTENHAM</td>\n",
       "      <td>19027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>79</td>\n",
       "      <td>103</td>\n",
       "      <td>115</td>\n",
       "      <td>193</td>\n",
       "      <td>142</td>\n",
       "      <td>119</td>\n",
       "      <td>71</td>\n",
       "      <td>89</td>\n",
       "      <td>30</td>\n",
       "      <td>9</td>\n",
       "      <td>...</td>\n",
       "      <td>375</td>\n",
       "      <td>STANBRIDGE ST &amp; W WASHINGTON ST</td>\n",
       "      <td>STANBRIDGE ST &amp; W WASHINGTON ST;  NORRISTOWN; ...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.117730</td>\n",
       "      <td>-75.357564</td>\n",
       "      <td>2016-01-04 02:05:06</td>\n",
       "      <td>EMS: FALL VICTIM</td>\n",
       "      <td>NORRISTOWN</td>\n",
       "      <td>19401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>69</td>\n",
       "      <td>60</td>\n",
       "      <td>53</td>\n",
       "      <td>92</td>\n",
       "      <td>171</td>\n",
       "      <td>167</td>\n",
       "      <td>123</td>\n",
       "      <td>80</td>\n",
       "      <td>198</td>\n",
       "      <td>646</td>\n",
       "      <td>...</td>\n",
       "      <td>2002</td>\n",
       "      <td>SHANNONDELL DR &amp; SHANNONDELL BLVD</td>\n",
       "      <td>SHANNONDELL DR &amp; SHANNONDELL BLVD;  LOWER PROV...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.133037</td>\n",
       "      <td>-75.408463</td>\n",
       "      <td>2016-01-04 01:59:27</td>\n",
       "      <td>EMS: SUBJECT IN PAIN</td>\n",
       "      <td>LOWER PROVIDENCE</td>\n",
       "      <td>19403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>57</td>\n",
       "      <td>55</td>\n",
       "      <td>64</td>\n",
       "      <td>171</td>\n",
       "      <td>147</td>\n",
       "      <td>167</td>\n",
       "      <td>137</td>\n",
       "      <td>51</td>\n",
       "      <td>84</td>\n",
       "      <td>153</td>\n",
       "      <td>...</td>\n",
       "      <td>1093</td>\n",
       "      <td>BROAD ST &amp; E 6TH ST</td>\n",
       "      <td>BROAD ST &amp; E 6TH ST;  LANSDALE; Station 345; 2...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.245673</td>\n",
       "      <td>-75.279667</td>\n",
       "      <td>2016-01-04 01:54:31</td>\n",
       "      <td>EMS: NAUSEA/VOMITING</td>\n",
       "      <td>LANSDALE</td>\n",
       "      <td>19446</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24</td>\n",
       "      <td>40</td>\n",
       "      <td>44</td>\n",
       "      <td>189</td>\n",
       "      <td>106</td>\n",
       "      <td>93</td>\n",
       "      <td>75</td>\n",
       "      <td>16</td>\n",
       "      <td>43</td>\n",
       "      <td>47</td>\n",
       "      <td>...</td>\n",
       "      <td>706</td>\n",
       "      <td>FAYETTE ST &amp; BUTLER PIKE</td>\n",
       "      <td>FAYETTE ST &amp; BUTLER PIKE;  PLYMOUTH; Station 3...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.082889</td>\n",
       "      <td>-75.297395</td>\n",
       "      <td>2016-01-04 01:55:14</td>\n",
       "      <td>EMS: FALL VICTIM</td>\n",
       "      <td>PLYMOUTH</td>\n",
       "      <td>19428</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 64 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   AGE_10_14  AGE_15_19  AGE_20_24  AGE_25_34  AGE_35_44  AGE_45_54  \\\n",
       "0         59         67         71        145        168        157   \n",
       "1         79        103        115        193        142        119   \n",
       "2         69         60         53         92        171        167   \n",
       "3         57         55         64        171        147        167   \n",
       "4         24         40         44        189        106         93   \n",
       "\n",
       "   AGE_55_64  AGE_5_9  AGE_65_74  AGE_75_84  ...    WHITE  \\\n",
       "0        244       63        232        257  ...     1286   \n",
       "1         71       89         30          9  ...      375   \n",
       "2        123       80        198        646  ...     2002   \n",
       "3        137       51         84        153  ...     1093   \n",
       "4         75       16         43         47  ...      706   \n",
       "\n",
       "                                addr  \\\n",
       "0               CEDAR RD & WARING RD   \n",
       "1    STANBRIDGE ST & W WASHINGTON ST   \n",
       "2  SHANNONDELL DR & SHANNONDELL BLVD   \n",
       "3                BROAD ST & E 6TH ST   \n",
       "4           FAYETTE ST & BUTLER PIKE   \n",
       "\n",
       "                                               desc_  e        lat        lng  \\\n",
       "0  CEDAR RD & WARING RD;  CHELTENHAM; Station 358...  1  40.071914 -75.118497   \n",
       "1  STANBRIDGE ST & W WASHINGTON ST;  NORRISTOWN; ...  1  40.117730 -75.357564   \n",
       "2  SHANNONDELL DR & SHANNONDELL BLVD;  LOWER PROV...  1  40.133037 -75.408463   \n",
       "3  BROAD ST & E 6TH ST;  LANSDALE; Station 345; 2...  1  40.245673 -75.279667   \n",
       "4  FAYETTE ST & BUTLER PIKE;  PLYMOUTH; Station 3...  1  40.082889 -75.297395   \n",
       "\n",
       "            timeStamp                 title               twp    zip  \n",
       "0 2016-01-04 02:22:12     EMS: HEMORRHAGING        CHELTENHAM  19027  \n",
       "1 2016-01-04 02:05:06      EMS: FALL VICTIM        NORRISTOWN  19401  \n",
       "2 2016-01-04 01:59:27  EMS: SUBJECT IN PAIN  LOWER PROVIDENCE  19403  \n",
       "3 2016-01-04 01:54:31  EMS: NAUSEA/VOMITING          LANSDALE  19446  \n",
       "4 2016-01-04 01:55:14      EMS: FALL VICTIM          PLYMOUTH  19428  \n",
       "\n",
       "[5 rows x 64 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sdf.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we want to parse timestamp to get more information from it.<br>\n",
    "We will extend the dataset with more time related values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdf['datetime'] = pd.to_datetime(sdf['timeStamp'], unit='ms')\n",
    "sdf['month'] = sdf['datetime'].dt.month\n",
    "sdf['day'] = sdf['datetime'].dt.day\n",
    "sdf['dayofweek'] = sdf['datetime'].dt.dayofweek\n",
    "sdf['week'] = sdf['datetime'].dt.week"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We want to group our dataframe by county FIPS and tract.<br>\n",
    "Framing a new column with combined code for county FIPS code and tract ID."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdf['cntytract'] = sdf['STCOFIPS'] + sdf.TRACT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    42091202301\n",
       "1    42091203804\n",
       "2    42091206006\n",
       "3    42091200901\n",
       "4    42091204101\n",
       "Name: cntytract, dtype: object"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sdf['cntytract'].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Montgomery County, locally also referred to as Montco, is a county located in the Commonwealth of Pennsylvania. As of the 2010 census, the population was 799,874, making it the third-most populous county in Pennsylvania, after Philadelphia and Allegheny Counties, and the 71st most populous in the United States.<br>\n",
    "So let's take pop2012 as one of the significant predictor for our analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_columns = [\"dayofweek\", \"week\", \"month\", \"day\", \"cntytract\", \"POP2012\", \"HOUSEHOLDS\",\n",
    "                 \"FAMILIES\", \"MED_AGE\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "df =  sdf[final_columns]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "groupby_list =   [\"dayofweek\", \"week\", \"month\", \"day\", \"cntytract\", \"POP2012\", \"HOUSEHOLDS\",\n",
    "                 \"FAMILIES\", \"MED_AGE\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdf_grouped = df.groupby(groupby_list).size().reset_index(name='count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "week         -0.015564\n",
       "dayofweek    -0.012391\n",
       "month        -0.009381\n",
       "day           0.006395\n",
       "MED_AGE       0.047343\n",
       "FAMILIES      0.111348\n",
       "POP2012       0.121887\n",
       "HOUSEHOLDS    0.138449\n",
       "count         1.000000\n",
       "Name: count, dtype: float64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_matrix = sdf_grouped.corr()\n",
    "corr_matrix['count'].sort_values()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "911 calls have more correlation with number of households and populaation in each block group.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = sdf_grouped['count']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "sdf_grouped.drop('count', inplace=True, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use sklearn library to split the dataframe into train and  test sets usin the **train_test split**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "train_X, test_X, train_y, test_y = train_test_split(sdf_grouped, y, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train and test your model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.10778108553494348"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.metrics import r2_score\n",
    "\n",
    "model = DecisionTreeRegressor(max_leaf_nodes=500)\n",
    "model.fit(train_X, train_y)\n",
    "predicted = model.predict(test_X)\n",
    "r2_score(test_y, predicted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.25730633396976876"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "forest_reg = RandomForestRegressor(n_estimators=200, min_samples_leaf=6, max_features=0.6, n_jobs=-1)\n",
    "forest = forest_reg.fit(train_X, train_y)\n",
    "Predicted = forest.predict(test_X)\n",
    "r2_score(test_y, Predicted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2632751482579958"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "validation_size = 0.20\n",
    "seed = 7\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(sdf_grouped,y,test_size = validation_size,random_state = seed)\n",
    "model = GradientBoostingRegressor(n_estimators=200, \n",
    "                                  learning_rate=0.1, max_depth=5, random_state=0, loss='ls', warm_start =  True)\n",
    "model.fit(X_train,Y_train)\n",
    "r2_score(Y_test, model.predict(X_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "oops! The model has poor predictions.  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's analyze our dataframe to understand what all factors could affect the number of 911 calls.\n",
    "Using the arcgis online organization you can navigate through the analysis variables availabe in the **enrich_layer** tools.\n",
    "We assume that the factors affecting number of 911 calls could be related to the number of smart phone buyers. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Enriching the **mont_block_groups** with demographic variables\n",
    "The enrich_layer tool gives you demographic and landascape data for the people, places, and businesses in a specific area, or within a selected travel time or distance from a location."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from arcgis.features.enrich_data import enrich_layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "blkgrp_smartphones_enrich = enrich_layer(mont_block_groups, \n",
    "                              analysis_variables=[\"PhonesYellowPages.MP27002A_B\"], \n",
    "                              output_name='blkgrp_smartphones_enrichfgbbgfhfgddfv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"item_container\" style=\"height: auto; overflow: hidden; border: 1px solid #cfcfcf; border-radius: 2px; background: #f6fafa; line-height: 1.21429em; padding: 10px;\">\n",
       "                    <div class=\"item_left\" style=\"width: 210px; float: left;\">\n",
       "                       <a href='https://deldev.maps.arcgis.com/home/item.html?id=779a829b099043858068b259680158d4' target='_blank'>\n",
       "                        <img src='http://static.arcgis.com/images/desktopapp.png' class=\"itemThumbnail\">\n",
       "                       </a>\n",
       "                    </div>\n",
       "\n",
       "                    <div class=\"item_right\"     style=\"float: none; width: auto; overflow: hidden;\">\n",
       "                        <a href='https://deldev.maps.arcgis.com/home/item.html?id=779a829b099043858068b259680158d4' target='_blank'><b>enrich__layer_with_1_variable</b>\n",
       "                        </a>\n",
       "                        <br/><img src='https://deldev.maps.arcgis.com/home/js/jsapi/esri/css/images/item_type_icons/featureshosted16.png' style=\"vertical-align:middle;\">Feature Layer Collection by demo_deldev\n",
       "                        <br/>Last Modified: June 16, 2018\n",
       "                        <br/>0 comments, 0 views\n",
       "                    </div>\n",
       "                </div>\n",
       "                "
      ],
      "text/plain": [
       "<Item title:\"enrich__layer_with_1_variable\" type:Feature Layer Collection owner:demo_deldev>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blkgrp_smartphones_enrich"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Again we use the **overlay_layers** tools from arcgis feature module.<br>\n",
    "Repeat the process to predict the mkodel on the test set.<br>\n",
    "Since we saw that month, week and dayofweek have negative as well as less correlations we are going to ignore those predictors\n",
    "and ue enriched variables as our predictors for our analysis. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "intersect_of_calls_with_grp = overlay_layers(mont_911_calls, blkgrp_smartphones_enrich, tolerance=0, context={},\n",
    "                           output_name=\"interect_911call_enrichedBlkGrp\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "en_sdf = pd.DataFrame.spatial.from_layer(intersect_of_calls_with_grp.layers[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(91495, 70)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "en_sdf.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['AGE_10_14', 'AGE_15_19', 'AGE_20_24', 'AGE_25_34', 'AGE_35_44',\n",
       "       'AGE_45_54', 'AGE_55_64', 'AGE_5_9', 'AGE_65_74', 'AGE_75_84',\n",
       "       'AGE_85_UP', 'AGE_UNDER5', 'AMERI_ES', 'ASIAN', 'AVE_FAM_SZ',\n",
       "       'AVE_HH_SZ', 'BLACK', 'BLKGRP', 'CNTY_FIPS', 'ENRICH_FID', 'FAMILIES',\n",
       "       'FEMALES', 'FHH_CHILD', 'FIPS', 'HAWN_PI', 'HISPANIC', 'HOUSEHOLDS',\n",
       "       'HSEHLD_1_F', 'HSEHLD_1_M', 'HSE_UNITS', 'HasData', 'ID', 'Join_Count',\n",
       "       'MALES', 'MARHH_CHD', 'MARHH_NO_C', 'MED_AGE', 'MED_AGE_F', 'MED_AGE_M',\n",
       "       'MHH_CHILD', 'MP27002a_B', 'MULT_RACE', 'OBJECTID_1', 'OTHER',\n",
       "       'OWNER_OCC', 'ObjectID', 'POP10_SQMI', 'POP12_SQMI', 'POP2010',\n",
       "       'POP2012', 'RENTER_OCC', 'SHAPE', 'SQMI', 'STATE_FIPS', 'STCOFIPS',\n",
       "       'TARGET_FID', 'TRACT', 'VACANT', 'WHITE', 'addr', 'aggregationMethod',\n",
       "       'desc_', 'e', 'lat', 'lng', 'sourceCountry', 'timeStamp', 'title',\n",
       "       'twp', 'zip'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "en_sdf.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "en_sdf['datetime'] = pd.to_datetime(en_sdf['timeStamp'], unit='ms')\n",
    "en_sdf['day'] = en_sdf['datetime'].dt.day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "en_sdf['cntytract'] = en_sdf['STCOFIPS'] + en_sdf.TRACT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols = [\"day\", \"cntytract\", \"POP2012\", \"HOUSEHOLDS\",\n",
    "                 \"FAMILIES\", \"MED_AGE\", \"MP27002a_B\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "en_df =  en_sdf[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "group_list =   [\"day\", \"cntytract\", \"POP2012\", \"HOUSEHOLDS\",\n",
    "                 \"FAMILIES\", \"MED_AGE\", \"MP27002a_B\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "en_sdf_grouped = en_df.groupby(group_list).size().reset_index(name='count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "day          -0.010811\n",
       "MED_AGE       0.077202\n",
       "FAMILIES      0.224070\n",
       "MP27002a_B    0.258259\n",
       "POP2012       0.261039\n",
       "HOUSEHOLDS    0.290274\n",
       "count         1.000000\n",
       "Name: count, dtype: float64"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_matrix = en_sdf_grouped.corr()\n",
    "corr_matrix['count'].sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "en_y = en_sdf_grouped['count']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "en_sdf_grouped.drop('count', inplace=True, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(en_sdf_grouped, en_y, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train and test your model to check improvement "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6898870637738551"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.metrics import r2_score\n",
    "\n",
    "model = DecisionTreeRegressor(max_leaf_nodes=500)\n",
    "model.fit(x_train, y_train)\n",
    "predicted = model.predict(x_test)\n",
    "r2_score(y_test, predicted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7510823142293463"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "forest_reg = RandomForestRegressor(n_estimators=200, min_samples_leaf=6, max_features=0.6, n_jobs=-1, oob_score=True)\n",
    "forest = forest_reg.fit(x_train, y_train)\n",
    "Predicted = forest.predict(x_test)\n",
    "r2_score(y_test, Predicted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7539521971735129"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "validation_size = 0.20\n",
    "seed = 7\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(en_sdf_grouped,en_y,test_size = validation_size,random_state = seed)\n",
    "model = GradientBoostingRegressor(n_estimators=200, \n",
    "                                  learning_rate=0.1, max_depth=5, random_state=0, loss='ls', warm_start =  True)\n",
    "model.fit(X_train,Y_train)\n",
    "r2_score(Y_test, model.predict(X_test))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Wow!! We have dramatically improved results now.\n",
    "By observing the category column of our dataframe we figure out that there are three major categories of 911 calls mainly traffic, fire and emergency.Pondering upon these as our factors we will now enrich our dataframe with relative road density, day time population of workers, unemployed population, and diversity index."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lets add more demographic variables using the enrich_layer tool."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "blkgrp_enrich = enrich_layer(blkgrp_smartphones_enrich, \n",
    "                              analysis_variables=[\"DaytimePopulation.DPOPWRK_CY\",\"industry.UNEMP_CY\",\n",
    "                                                  \"KeyUSFacts.DIVINDX_CY\",\"Infrastructure.ROADSUM\"], \n",
    "                              output_name='enrich__layer_with_5more_variables')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "intersect_of_enriched_blockgrp_calls = overlay_layers(mont_911_calls, blkgrp_enrich, tolerance=0, context={},\n",
    "                           output_name=\"enriched\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "edf = pd.DataFrame.spatial.from_layer(intersect_of_enriched_blockgrp_calls.layers[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "edf['category'],edf['category2'] = edf['title'].str.split(':',1).str"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "edf = edf.drop(['title','category2'],axis = 1)\n",
    "cat_unique = pd.DataFrame(edf.category.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>AGE_10_14</th>\n",
       "      <th>AGE_15_19</th>\n",
       "      <th>AGE_20_24</th>\n",
       "      <th>AGE_25_34</th>\n",
       "      <th>AGE_35_44</th>\n",
       "      <th>AGE_45_54</th>\n",
       "      <th>AGE_55_64</th>\n",
       "      <th>AGE_5_9</th>\n",
       "      <th>AGE_65_74</th>\n",
       "      <th>AGE_75_84</th>\n",
       "      <th>...</th>\n",
       "      <th>desc_</th>\n",
       "      <th>e</th>\n",
       "      <th>lat</th>\n",
       "      <th>lng</th>\n",
       "      <th>sourceCountry</th>\n",
       "      <th>sourceCountry_1</th>\n",
       "      <th>timeStamp</th>\n",
       "      <th>twp</th>\n",
       "      <th>zip</th>\n",
       "      <th>category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>20</td>\n",
       "      <td>23</td>\n",
       "      <td>29</td>\n",
       "      <td>118</td>\n",
       "      <td>86</td>\n",
       "      <td>116</td>\n",
       "      <td>95</td>\n",
       "      <td>20</td>\n",
       "      <td>66</td>\n",
       "      <td>16</td>\n",
       "      <td>...</td>\n",
       "      <td>S SCHUYLKILL AVE;  WEST NORRITON; Station 308B...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.123710</td>\n",
       "      <td>-75.382390</td>\n",
       "      <td>US</td>\n",
       "      <td>US</td>\n",
       "      <td>2016-01-01 08:05:57</td>\n",
       "      <td>WEST NORRITON</td>\n",
       "      <td>19403</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10</td>\n",
       "      <td>14</td>\n",
       "      <td>61</td>\n",
       "      <td>302</td>\n",
       "      <td>87</td>\n",
       "      <td>51</td>\n",
       "      <td>24</td>\n",
       "      <td>12</td>\n",
       "      <td>16</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>GULPH RD &amp; MALL BLVD;  UPPER MERION; Station 3...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.085338</td>\n",
       "      <td>-75.396135</td>\n",
       "      <td>US</td>\n",
       "      <td>US</td>\n",
       "      <td>2016-01-01 08:09:36</td>\n",
       "      <td>UPPER MERION</td>\n",
       "      <td>19406</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>104</td>\n",
       "      <td>117</td>\n",
       "      <td>82</td>\n",
       "      <td>233</td>\n",
       "      <td>275</td>\n",
       "      <td>316</td>\n",
       "      <td>299</td>\n",
       "      <td>104</td>\n",
       "      <td>163</td>\n",
       "      <td>283</td>\n",
       "      <td>...</td>\n",
       "      <td>BRITTANY PT &amp; VALLEY FORGE RD;  UPPER GWYNEDD;...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.229961</td>\n",
       "      <td>-75.313855</td>\n",
       "      <td>US</td>\n",
       "      <td>US</td>\n",
       "      <td>2016-01-01 08:25:33</td>\n",
       "      <td>UPPER GWYNEDD</td>\n",
       "      <td>19446</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>127</td>\n",
       "      <td>116</td>\n",
       "      <td>139</td>\n",
       "      <td>229</td>\n",
       "      <td>203</td>\n",
       "      <td>354</td>\n",
       "      <td>299</td>\n",
       "      <td>97</td>\n",
       "      <td>160</td>\n",
       "      <td>61</td>\n",
       "      <td>...</td>\n",
       "      <td>OAKWOOD DR &amp; JASON DR;  UPPER MORELAND; Statio...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.166612</td>\n",
       "      <td>-75.067830</td>\n",
       "      <td>US</td>\n",
       "      <td>US</td>\n",
       "      <td>2016-01-01 08:26:09</td>\n",
       "      <td>UPPER MORELAND</td>\n",
       "      <td>19006</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>59</td>\n",
       "      <td>58</td>\n",
       "      <td>109</td>\n",
       "      <td>203</td>\n",
       "      <td>123</td>\n",
       "      <td>99</td>\n",
       "      <td>68</td>\n",
       "      <td>83</td>\n",
       "      <td>34</td>\n",
       "      <td>19</td>\n",
       "      <td>...</td>\n",
       "      <td>MAIN ST &amp; GREEN ST;  NORRISTOWN; Station 308A;...</td>\n",
       "      <td>1</td>\n",
       "      <td>40.113372</td>\n",
       "      <td>-75.341397</td>\n",
       "      <td>US</td>\n",
       "      <td>US</td>\n",
       "      <td>2016-01-01 08:27:05</td>\n",
       "      <td>NORRISTOWN</td>\n",
       "      <td>19401</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 79 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   AGE_10_14  AGE_15_19  AGE_20_24  AGE_25_34  AGE_35_44  AGE_45_54  \\\n",
       "0         20         23         29        118         86        116   \n",
       "1         10         14         61        302         87         51   \n",
       "2        104        117         82        233        275        316   \n",
       "3        127        116        139        229        203        354   \n",
       "4         59         58        109        203        123         99   \n",
       "\n",
       "   AGE_55_64  AGE_5_9  AGE_65_74  AGE_75_84    ...     \\\n",
       "0         95       20         66         16    ...      \n",
       "1         24       12         16          8    ...      \n",
       "2        299      104        163        283    ...      \n",
       "3        299       97        160         61    ...      \n",
       "4         68       83         34         19    ...      \n",
       "\n",
       "                                               desc_  e        lat        lng  \\\n",
       "0  S SCHUYLKILL AVE;  WEST NORRITON; Station 308B...  1  40.123710 -75.382390   \n",
       "1  GULPH RD & MALL BLVD;  UPPER MERION; Station 3...  1  40.085338 -75.396135   \n",
       "2  BRITTANY PT & VALLEY FORGE RD;  UPPER GWYNEDD;...  1  40.229961 -75.313855   \n",
       "3  OAKWOOD DR & JASON DR;  UPPER MORELAND; Statio...  1  40.166612 -75.067830   \n",
       "4  MAIN ST & GREEN ST;  NORRISTOWN; Station 308A;...  1  40.113372 -75.341397   \n",
       "\n",
       "   sourceCountry  sourceCountry_1           timeStamp             twp    zip  \\\n",
       "0             US               US 2016-01-01 08:05:57   WEST NORRITON  19403   \n",
       "1             US               US 2016-01-01 08:09:36    UPPER MERION  19406   \n",
       "2             US               US 2016-01-01 08:25:33   UPPER GWYNEDD  19446   \n",
       "3             US               US 2016-01-01 08:26:09  UPPER MORELAND  19006   \n",
       "4             US               US 2016-01-01 08:27:05      NORRISTOWN  19401   \n",
       "\n",
       "   category  \n",
       "0         0  \n",
       "1         0  \n",
       "2         0  \n",
       "3         0  \n",
       "4         0  \n",
       "\n",
       "[5 rows x 79 columns]"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 0 = EMS (Emergency Medical Services)\n",
    "#1 = FIRE\n",
    "# 2 =  TRAFFIC\n",
    "CATEGORIES = {'EMS':0,'Fire':1,'Traffic':2}\n",
    "edf['category'].replace(CATEGORIES,inplace=True)\n",
    "edf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "edf['datetime'] = pd.to_datetime(edf['timeStamp'], unit='ms')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "edf['day'] = edf['datetime'].dt.day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "edf['cntytract'] = edf['STCOFIPS'] + edf.TRACT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols = [\"day\", \"cntytract\", \"POP2012\", \"HOUSEHOLDS\", \"category\",\n",
    "                 \"FAMILIES\", \"MED_AGE\", \"MP27002a_B\", \"DIVINDX_CY\", \"DPOPWRK_CY\", \"RoadSum\", \"UNEMP_CY\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "rdf =  edf[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "group_list = [\"day\", \"cntytract\", \"POP2012\", \"HOUSEHOLDS\",\n",
    "                 \"FAMILIES\", \"MED_AGE\", \"MP27002a_B\", \"DIVINDX_CY\", \"DPOPWRK_CY\", \"RoadSum\", \"UNEMP_CY\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "edf_grouped = rdf.groupby(group_list).size().reset_index(name='count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "vdf = edf_grouped.copy() # making a copy for use in visualization later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "day          -0.010811\n",
       "DIVINDX_CY    0.060770\n",
       "MED_AGE       0.077202\n",
       "UNEMP_CY      0.119524\n",
       "FAMILIES      0.224070\n",
       "MP27002a_B    0.258259\n",
       "POP2012       0.261039\n",
       "HOUSEHOLDS    0.290274\n",
       "RoadSum       0.306760\n",
       "DPOPWRK_CY    0.400352\n",
       "count         1.000000\n",
       "Name: count, dtype: float64"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_matrix = edf_grouped.corr()\n",
    "corr_matrix['count'].sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "enriched_y = edf_grouped['count']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "edf_grouped.drop('count', inplace=True, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(edf_grouped, enriched_y, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Results after enriching data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6970121511890337"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "\n",
    "model = DecisionTreeRegressor(max_leaf_nodes=500)\n",
    "model.fit(x_train, y_train)\n",
    "predicted = model.predict(x_test)\n",
    "r2_score(y_test, predicted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7527476647519262"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "forest_reg = RandomForestRegressor(n_estimators=200, min_samples_leaf=6, max_features=0.6, n_jobs=-1, oob_score=True)\n",
    "forest = forest_reg.fit(x_train, y_train)\n",
    "Predicted = forest.predict(x_test)\n",
    "r2_score(y_test, Predicted)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7638603774109038"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "validation_size = 0.20\n",
    "seed = 7\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(edf_grouped,enriched_y,test_size = validation_size,random_state = seed)\n",
    "model = GradientBoostingRegressor(n_estimators=200, \n",
    "                                  learning_rate=0.1, max_depth=5, random_state=0, loss='ls', warm_start =  True)\n",
    "model.fit(X_train,Y_train)\n",
    "r2_score(Y_test, model.predict(X_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we see that day time population of workers is closely related to the number of 911 calls, we have got slightly improved results.\n",
    "Wow!! That gives me motivation to dig deep into demographic variables that can dramatically improve the performance  of our model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.matrix.ClusterGrid at 0x2388d57fbe0>"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2388d514c88>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.clustermap(vdf.corr())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The cluster map explains that the count for 911 calls is the most correlated to DPOPWRK_CY.<br>\n",
    "The color matrix of the \"count\" row shows that 911 calls are correlated to 'RoadSum', 'POP2012', 'MP27002a_B', 'FAMILIES' in the range of 0.3-0.4.<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}