Visualization.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime\n",
    "import pandas as pd\n",
    "import json\n",
    "import os\n",
    "\n",
    "with open('data/overlap/overlaps.json', 'r', encoding='UTF-8') as f:\n",
    "    overlaps = json.load(f)\n",
    "    \n",
    "novel_words = {}\n",
    "\n",
    "for file in overlaps:\n",
    "    novel_words[file] = []\n",
    "    for el in overlaps[file]:\n",
    "        if el in ['h', 't']:\n",
    "            for ngram in overlaps[file][el]:\n",
    "                if isinstance(overlaps[file][el][ngram], float):\n",
    "                    novel_words[file].append(overlaps[file][el][ngram])\n",
    "                else:\n",
    "                    novel_words[file].append(None)\n",
    "            \n",
    "columns = ['h_novel_1', 'h_novel_2', 'h_novel_3', 'h_novel_4', \\\n",
    "           't_novel_1', 't_novel_2', 't_novel_3', 't_novel_4']\n",
    "\n",
    "df = pd.DataFrame.from_dict(novel_words, orient='index', columns=columns)\n",
    "\n",
    "article_len = []\n",
    "heading_len = []\n",
    "title_len = []\n",
    "dates = []\n",
    "for idx, file in enumerate(overlaps):\n",
    "    with open(os.path.join('data/docs/parsed', file), 'r', encoding='UTF-8') as f:\n",
    "        doc = json.load(f)\n",
    "    article_len.append(len(doc['article'].split()))\n",
    "    heading_len.append(len(doc['heading'].split()))\n",
    "    title_len.append(len(doc['title'].split()))\n",
    "    dates.append(datetime.datetime.strptime(doc['date'], '%Y-%m-%d'))\n",
    "        \n",
    "df['article_len'] = article_len\n",
    "df['heading_len'] = heading_len\n",
    "df['title_len'] = title_len\n",
    "df['dates']= dates\n",
    "df['year'] = pd.DatetimeIndex(df['dates']).year\n",
    "df['heading_article_len_ratio'] = df['heading_len'] / df['article_len']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>h_novel_1</th>\n",
       "      <th>h_novel_2</th>\n",
       "      <th>h_novel_3</th>\n",
       "      <th>h_novel_4</th>\n",
       "      <th>t_novel_1</th>\n",
       "      <th>t_novel_2</th>\n",
       "      <th>t_novel_3</th>\n",
       "      <th>t_novel_4</th>\n",
       "      <th>article_len</th>\n",
       "      <th>heading_len</th>\n",
       "      <th>title_len</th>\n",
       "      <th>dates</th>\n",
       "      <th>year</th>\n",
       "      <th>heading_article_len_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10.json</th>\n",
       "      <td>35.5</td>\n",
       "      <td>87.1</td>\n",
       "      <td>93.3</td>\n",
       "      <td>96.6</td>\n",
       "      <td>33.3</td>\n",
       "      <td>81.8</td>\n",
       "      <td>90.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>539</td>\n",
       "      <td>32</td>\n",
       "      <td>12</td>\n",
       "      <td>2020-07-02</td>\n",
       "      <td>2020</td>\n",
       "      <td>0.059369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1000.json</th>\n",
       "      <td>25.9</td>\n",
       "      <td>58.1</td>\n",
       "      <td>83.9</td>\n",
       "      <td>93.3</td>\n",
       "      <td>20.0</td>\n",
       "      <td>60.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>291</td>\n",
       "      <td>33</td>\n",
       "      <td>11</td>\n",
       "      <td>2020-06-15</td>\n",
       "      <td>2020</td>\n",
       "      <td>0.113402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10000.json</th>\n",
       "      <td>19.2</td>\n",
       "      <td>40.7</td>\n",
       "      <td>53.8</td>\n",
       "      <td>68.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>109</td>\n",
       "      <td>28</td>\n",
       "      <td>6</td>\n",
       "      <td>2017-11-18</td>\n",
       "      <td>2017</td>\n",
       "      <td>0.256881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10001.json</th>\n",
       "      <td>50.0</td>\n",
       "      <td>91.7</td>\n",
       "      <td>100.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>12.5</td>\n",
       "      <td>62.5</td>\n",
       "      <td>85.7</td>\n",
       "      <td>100.0</td>\n",
       "      <td>170</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>2017-11-22</td>\n",
       "      <td>2017</td>\n",
       "      <td>0.076471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10002.json</th>\n",
       "      <td>100.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>38.5</td>\n",
       "      <td>66.7</td>\n",
       "      <td>81.8</td>\n",
       "      <td>90.0</td>\n",
       "      <td>234</td>\n",
       "      <td>2</td>\n",
       "      <td>15</td>\n",
       "      <td>2018-08-07</td>\n",
       "      <td>2018</td>\n",
       "      <td>0.008547</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            h_novel_1  h_novel_2  h_novel_3  h_novel_4  t_novel_1  t_novel_2  \\\n",
       "10.json          35.5       87.1       93.3       96.6       33.3       81.8   \n",
       "1000.json        25.9       58.1       83.9       93.3       20.0       60.0   \n",
       "10000.json       19.2       40.7       53.8       68.0       20.0       75.0   \n",
       "10001.json       50.0       91.7      100.0      100.0       12.5       62.5   \n",
       "10002.json      100.0      100.0        NaN        NaN       38.5       66.7   \n",
       "\n",
       "            t_novel_3  t_novel_4  article_len  heading_len  title_len  \\\n",
       "10.json          90.0      100.0          539           32         12   \n",
       "1000.json       100.0      100.0          291           33         11   \n",
       "10000.json      100.0      100.0          109           28          6   \n",
       "10001.json       85.7      100.0          170           13         10   \n",
       "10002.json       81.8       90.0          234            2         15   \n",
       "\n",
       "                dates  year  heading_article_len_ratio  \n",
       "10.json    2020-07-02  2020                   0.059369  \n",
       "1000.json  2020-06-15  2020                   0.113402  \n",
       "10000.json 2017-11-18  2017                   0.256881  \n",
       "10001.json 2017-11-22  2017                   0.076471  \n",
       "10002.json 2018-08-07  2018                   0.008547  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6529\n"
     ]
    }
   ],
   "source": [
    "print(len(df.loc[df['heading_len'] < 5])) # too many, let's discard them\n",
    "df = df.loc[df['heading_len'] >= 5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>h_novel_1</th>\n",
       "      <th>h_novel_2</th>\n",
       "      <th>heading_len</th>\n",
       "      <th>heading_article_len_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>h_novel_1</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.842821</td>\n",
       "      <td>0.288471</td>\n",
       "      <td>0.493383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h_novel_2</th>\n",
       "      <td>0.842821</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.248969</td>\n",
       "      <td>0.289449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>heading_len</th>\n",
       "      <td>0.288471</td>\n",
       "      <td>0.248969</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.487751</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>heading_article_len_ratio</th>\n",
       "      <td>0.493383</td>\n",
       "      <td>0.289449</td>\n",
       "      <td>0.487751</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           h_novel_1  h_novel_2  heading_len  \\\n",
       "h_novel_1                   1.000000   0.842821     0.288471   \n",
       "h_novel_2                   0.842821   1.000000     0.248969   \n",
       "heading_len                 0.288471   0.248969     1.000000   \n",
       "heading_article_len_ratio   0.493383   0.289449     0.487751   \n",
       "\n",
       "                           heading_article_len_ratio  \n",
       "h_novel_1                                   0.493383  \n",
       "h_novel_2                                   0.289449  \n",
       "heading_len                                 0.487751  \n",
       "heading_article_len_ratio                   1.000000  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['h_novel_1', 'h_novel_2', 'heading_len','heading_article_len_ratio']].corr()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Important correlation between h_novel_1 and heading_article_len_ratio"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### __% novel ngrams Distribution__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fc922e6d4c0>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.set()\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "plt.figure(figsize=(20,5))\n",
    "sns.histplot(data=df, x=\"h_novel_1\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### __Distribution shift over years__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJkAAAFSCAYAAABR3Q77AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3dfZCdZX0//vdu9oHEJG4SVlySaIZI0iiKmLVYf4WUBCS0hIdxMjApOKVgq2KM0AQoxORrQqQhmeGhBsNU60w1A1MHTUxAlloEa+1XRMSCgCIgYhKBJAQSEvbx/P7w2y2RPGw45+w5u/t6zTDDue+zn/0cdq8993lzXdddUygUCgEAAACAItRWugEAAAAABj4hEwAAAABFEzIBAAAAUDQhEwAAAABFEzIBAAAAUDQhEwAAAABFEzIBAAAAUDQhEwBAiXR0dOTqq6/OKaeckhNOOCHnnHNO7r///t7z//Vf/5XZs2fn+OOPz4UXXpjNmzf3nrvrrrty/vnn9557vQcffDAnnHDCPv9MnTo1bW1t/fbaAAAORcgEAFAiXV1daWlpyde+9rX85Cc/yYIFC/LZz342v/3tb7Njx458+tOfzoIFC/LAAw/kuOOOy2WXXdb7tU1NTfnYxz6Wj3/842+o29ramp/+9Ke9/6xduzYjRozISSed1J8vDwDgoOoq3QAAwGAxYsSIzJ8/v/fxKaeckgkTJuTnP/95du7cmWOPPTZnnHFGkmT+/Pn50Ic+lKeeeiqTJ0/Ohz/84STJN77xjUN+n/Xr12f27NkZMWJEeV4IAMCbYCYTAECZbNu2Lb/+9a/zrne9K08++WSmTp3ae27EiBF5xzvekV/96leHVXPv3r25++67c84555S6XQCAogiZAADKoLOzMwsXLsy5556byZMnZ8+ePRk1atQ+zxk5cmReffXVw6rb1taWMWPG5I//+I9L2S4AQNGETAAAJdbT05Mrrrgi9fX1+dznPpfk9zOXdu/evc/zXn311bzlLW85rNrr16/POeeck5qampL1CwBQCkImAIASKhQKueaaa7Jt27b84z/+Y+rr65Mkxx57bJ544one5+3Zsye/+c1v8q53vavPtbdu3ZoHHnjAUjkAoCoJmQAASmjp0qV56qmnsnbt2hxxxBG9x0877bQ8+eSTaWtrS3t7e9asWZOpU6dm8uTJSZLu7u60t7enq6srPT09aW9vT2dn5z61N2zYkBNOOCHveMc7+vU1AQD0RU2hUChUugkAgMFg8+bNmTlzZhoaGlJX97838f385z+fs846Kz/84Q+zbNmybNmyJccff3yuu+66TJgwIUnyzW9+M3//93+/T71zzz03//AP/9D7ePbs2bn44oszd+7c/nlBAACHQcgEAAAAQNEslwMAAACgaEImAAAAAIomZAIAAACgaEImAAAAAIomZAIAAACgaEImAAAAAIpWV+kGyu2ll15NT0+h0m0UZdy4kdm+fXel24CqZ6xA3xgr0DfGCvSNsQJ9M1jGSm1tTcaMect+zw36kKmnpzDgQ6Ykg+I1QH8wVqBvjBXoG2MF+sZYgb4Z7GPFcjkAAAAAiiZkAgAAAKBoQiYAAAAAiiZkAgAAAKBoQiYAAAAAiiZkAgAAAKBoQiYAAAAAiiZkAgAAAKBoQiYAAAAAilZX6QYAAEhGv3V4GhtKd2nW3tGVV17eW7J6AACHImQCAKgCjQ11WXTT/SWrt2rBjJLVAgDoC8vlAAAAACiakAkAAACAogmZAAAAACiakAkAAACAogmZAAAAACiakAkAAACAogmZAAAAACiakAkAAACAogmZAAAAAChaXX99o5UrV6atrS2bN2/Oxo0bM2XKlPz2t7/NpZde2vucXbt2Zffu3XnggQeSJDNnzkxDQ0MaGxuTJAsXLsxJJ53UXy0DAAAA0Ef9FjLNmjUrH/vYx/KXf/mXvccmTJiQDRs29D5esWJFuru79/m6m2++OVOmTOmvNgEAAAB4E/otZGptbT3o+Y6OjmzcuDFf+cpX+qkjAAAAAEql30KmQ7n33ntz1FFH5T3vec8+xxcuXJhCoZDp06fn8ssvz+jRow+r7rhxI0vZZsU0N4+qdAswIBgr0DfGSnWqry/tpZmfc/H8N4S+MVagbwb7WKmakOmOO+7IRz/60X2OrVu3Li0tLeno6MiKFSuybNmyrF69+rDqbt++Oz09hVK22u+am0flxRd3VboNqHrGCvSNsVKdmptHpbOzq6Q1/ZyLY6xA3xgr0DeDZazU1tYccEJPVdxd7vnnn8+Pf/zjzJkzZ5/jLS0tSZKGhobMmzcvDz30UCXaAwAAAOAQqiJk+ta3vpUZM2ZkzJgxvcf27NmTXbt+n/AVCoXcddddmTZtWqVaBAAAAOAg+m253LXXXpt77rkn27Zty0UXXZSmpqbceeedSX4fMl1zzTX7PH/79u2ZP39+uru709PTk8mTJ2fp0qX91S4AAAAAh6HfQqbFixdn8eLF+z3X1tb2hmMTJ07M+vXry90WAAAAACVQFcvlAAAAABjYhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFE3IBAAAAEDRhEwAAAAAFK2uv77RypUr09bWls2bN2fjxo2ZMmVKkmTmzJlpaGhIY2NjkmThwoU56aSTkiQPP/xwlixZkvb29owfPz6rVq3KuHHj+qtlAAAAAPqo32YyzZo1K+vWrcv48ePfcO7mm2/Ohg0bsmHDht6AqVAoZNGiRVmyZEna2trS2tqa1atX91e7AAAAAByGfguZWltb09LS0ufnP/LII2lsbExra2uS5Pzzz8/dd99drvYAAAAAKEK/LZc7mIULF6ZQKGT69Om5/PLLM3r06GzdujVHH31073PGjh2bnp6e7Ny5M01NTX2uPW7cyHK03O+am0dVugUYEIwV6BtjpTrV15f20szPuXj+G0LfGCvQN4N9rFQ8ZFq3bl1aWlrS0dGRFStWZNmyZSVdFrd9++709BRKVq8SmptH5cUXd1W6Dah6xgr0jbFSnZqbR6Wzs6ukNf2ci2OsQN8YK9A3g2Ws1NbWHHBCT8XvLvc/S+gaGhoyb968PPTQQ73Ht2zZ0vu8HTt2pKam5rBmMQEAAADQPyoaMu3Zsye7dv0+xSsUCrnrrrsybdq0JMlxxx2X1157LQ8++GCS5Pbbb88ZZ5xRsV4BAAAAOLB+Wy537bXX5p577sm2bdty0UUXpampKWvXrs38+fPT3d2dnp6eTJ48OUuXLk2S1NbW5vrrr8/SpUvT3t6e8ePHZ9WqVf3VLgAAAACHod9CpsWLF2fx4sVvOL5+/foDfs0HPvCBbNy4sZxtAQAAAFACFd+TCQAAAICBT8gEAAAAQNGETAAAAAAUTcgEAAAAQNGETAAAAAAUTcgEAAAAQNGETAAAAAAUTcgEAAAAQNGETAAAAAAUTcgEAAAAQNHqKt0AAABUo86unjQ3jyphve7U1w0rWb0kae/oyisv7y1pTQB4s4RMAACwH/V1tVl00/0lq7dqwYyS1vufmgBQLSyXAwAAAKBoQiYAAAAAiiZkAgAAAKBo9mQCAKAiRr91eBobSnc5ahNsAKgsIRMAABXR2FBX8o21AYDKsVwOAAAAgKIJmQAAAAAompAJAAAAgKIJmQAAAAAompAJAAAAgKIJmQAAAAAompAJAAAAgKIJmQAAAAAoWl2lGwAAoPQ6u3rS3DyqZPXaO7ryyst7S1YPABh8+i1kWrlyZdra2rJ58+Zs3LgxU6ZMyUsvvZQrrrgiv/nNb9LQ0JB3vvOdWbZsWcaOHZskmTp1aqZMmZLa2t9PuLr++uszderU/moZAGDAqq+rzaKb7i9ZvVULZpSsFgAwOPXbcrlZs2Zl3bp1GT9+fO+xmpqaXHLJJWlra8vGjRszceLErF69ep+vu/3227Nhw4Zs2LBBwAQAAABQpfotZGptbU1LS8s+x5qamnLiiSf2Pn7/+9+fLVu29FdLAAAAAJRI1ezJ1NPTk9tuuy0zZ87c5/iFF16Y7u7unHzyyZk/f34aGhoOq+64cSNL2WbFlHJPBRjMjBXoG2OlOtXXl/bSrNT1yvF7U+09VvvPJDGeqQ5+D6FvBvtYqZqQafny5RkxYkQuuOCC3mP33XdfWlpasnv37ixatChr1qzJZZdddlh1t2/fnZ6eQqnb7VfNzaPy4ou7Kt0GVD1jBfrGWKlOzc2j0tnZVdKapa5X6t+bcrzmUvY4EH4mSel/LnC4vK9A3wyWsVJbW3PACT39tlzuYFauXJlnn302N954Y+8m30l6l9eNHDkyc+fOzUMPPVSpFgEAAAA4iIqHTDfccEMeffTRrFmzZp+lcC+//HJee+21JElXV1fa2toybdq0SrUJAAAAwEH023K5a6+9Nvfcc0+2bduWiy66KE1NTbnxxhuzdu3aTJo0Keeff36SZMKECVmzZk2efvrpLFmyJDU1Nenq6soJJ5yQBQsW9Fe7AAAAAByGfguZFi9enMWLF7/h+C9+8Yv9Pv+EE07Ixo0by90WAAAAACVQNRt/AwAAh6ezq6ekdypq7+jKKy/vLVk9AIYWIRMAAAxQ9XW1WXTT/SWrt2rBjJLVAmDoqfjG3wAAAAAMfEImAAAAAIomZAIAAACgaEImAAAAAIomZAIAAACgaO4uBwDAoNDZ1ZPm5lGVbgMAhiwhEwAAhzQQApz6utosuun+ktVbtWBGyWoBwFAgZAIA4JBKHeAkQhwAGGzsyQQAAABA0YRMAAAAABRNyAQAAABA0YRMAAAAABRNyAQAAABA0YRMAAAAABRNyAQAAABA0YRMAAAAABStzyHTd77znf0ev/vuu0vWDAAAAAADU59DpmuuuWa/x5csWVKyZgAAAAAYmOoO9YTnnnsuSVIoFHr//fXnGhoaytMZAAAAAAPGIUOm0047LTU1NSkUCjnttNP2OXfkkUdm/vz5ZWsOAAAAgIHhkCHTE088kSS54IIL8vWvf73sDQEAAAAw8PR5TyYBEwAAAAAHcsiZTP/jueeey4033pjHH388e/bs2efcfffdV+q+AAAAABhA+hwyLVy4MBMnTsyVV16Z4cOHl7MnAAAAAAaYPodMTz75ZG677bbU1vZ5hV2vlStXpq2tLZs3b87GjRszZcqUJMkzzzyTq666Kjt37kxTU1NWrlyZSZMmHfIcAAAAANWlz4nRBz/4wTz22GNv6pvMmjUr69aty/jx4/c5vnTp0sybNy9tbW2ZN29elixZ0qdzAAAAAFSXPs9kGj9+fC6++OJ85CMfyZFHHrnPuQULFhz0a1tbW99wbPv27Xnsscfy1a9+NUly5plnZvny5dmxY0cKhcIBz40dO7avLQMAAADQT/ocMu3duzczZ85MV1dXfve73xX9jbdu3Zqjjjoqw4YNS5IMGzYsb3vb27J169YUCoUDnjvckGncuJFF91oNmptHVboFGBCMFegbY6U61df3+dJsUNQrR82hVq8cNf194M3wewN9M9jHSp/fka677rpy9lE227fvTk9PodJtFKW5eVRefHFXpduAqmesQN8YK9WpuXlUOju7Slqz2uuVo+ZQq1eOmv4+cLi8r0DfDJaxUltbc8AJPX0OmZ577rkDnps4ceJhN9XS0pLnn38+3d3dGTZsWLq7u/PCCy+kpaUlhULhgOcAAAAAqD59DplOO+201NTUpFD431lBNTU1SZLHH3/8sL/xuHHjMm3atGzatClnn312Nm3alGnTpvUuhzvYOQAAAACqS59DpieeeGKfxy+++GK++MUv7ndT7z907bXX5p577sm2bdty0UUXpampKXfeeWf+z//5P7nqqqtyyy23ZPTo0Vm5cmXv1xzsHAAAAADV5U3vEtjc3Jxrrrkmp59+eubMmXPQ5y5evDiLFy9+w/HJkyfnG9/4xn6/5mDnAAAAAKgutcV88dNPP529e/eWqhcAAAAABqg+z2SaN29e7x5MSbJ379786le/yqWXXlqWxgCAgWP0W4ensaF0t1Fv7+jKKy/7H1kAAANJn68G586du8/j4cOH54/+6I8yadKkUvcEAAwwjQ11WXTT/SWrt2rBjJLVAgCgf/Q5ZDr33HPL2QcAAAAAA1if92Tq7OzMzTffnFmzZuW9731vZs2alZtvvjkdHR3l7A8AAACAAaDPM5lWrVqV//7v/87nP//5HH300dmyZUtuueWW7N69O1dffXU5ewQAAACgyvU5ZLr77ruzYcOGjBkzJklyzDHH5N3vfnfOPvtsIRMAAAwCnV09aW4eVbJ6NvEHGFr6HDIVCoXDOg4AAAws9XW1NvEH4E3r855Ms2fPzic/+cn8x3/8R5566ql8//vfz6WXXprZs2eXsz8AAAAABoA+z2RatGhRvvSlL2XZsmV54YUXctRRR+Uv/uIv8slPfrKc/QEAAAAwABxyJtNPfvKTrFq1Kg0NDVmwYEH+7d/+LT/72c9yzz33pKOjI4899lh/9AkAAABAFTtkyHTrrbfmgx/84H7PnXjiiVm7dm3JmwIAAABgYDlkyPT444/npJNO2u+5D3/4w3n00UdL3hQAAAAAA8shQ6bdu3ens7Nzv+e6urry6quvlrwpAAAAAAaWQ278fcwxx+QHP/hBTj311Dec+8EPfpBjjjmmLI0BAENXZ1dPmptHlaxee0dXXnl5b8nqAQDwRocMmf7qr/4qS5cuTU9PT0499dTU1tamp6cn3/3ud7Ns2bJcddVV/dEnADCE1NfVZtFN95es3hcuPUloBQBQZocMmebMmZNt27blyiuvTGdnZ5qamrJz5840NDTkM5/5TM4888z+6BMA4E0rdWi1asGMktUCABgsDhkyJclFF12UuXPn5qc//Wl27tyZpqamnHDCCRk5cmS5+wMAAABgAOhTyJQkI0eOPOBd5gAAAAAY2g55dzkAAAAAOBQhEwAAAABF6/NyOQCgb0a/dXgaG0r3FutOZgAADARCJgAoscaGOncyAwBgyLFcDgAAAICiCZkAAAAAKJqQCQAAAICiVXxPpt/+9re59NJLex/v2rUru3fvzgMPPJCZM2emoaEhjY2NSZKFCxfmpJNOqlSrAAAAABxAxUOmCRMmZMOGDb2PV6xYke7u7t7HN998c6ZMmVKJ1gBgUOrs6klz86hKtwEAwCBT8ZDp9To6OrJx48Z85StfqXQrADBo1dfVlvTud4k74AEAUGUh07333pujjjoq73nPe3qPLVy4MIVCIdOnT8/ll1+e0aNHH1bNceNGlrrNivB/nKFvjBWqRX19ad9iS/27Xer+ylGz2uuV4+9Ntb9mvzfVV68cNUtZrxwzJzu7elJfZ2vZauMaDPpmsI+VqgqZ7rjjjnz0ox/tfbxu3bq0tLSko6MjK1asyLJly7J69erDqrl9++709BRK3Wq/am4elRdf3FXpNqDqGStUi+bmUens7CppzVL+bpejvyQlr1nt9Ur996YcP5dqr1eOmkOtXjlqlrJeuWZOer+vLq7BoG8Gy1ipra054ISeqvlfAM8//3x+/OMfZ86cOb3HWlpakiQNDQ2ZN29eHnrooUq1BwAAAMBBVE3I9K1vfSszZszImDFjkiR79uzJrl2/T/gKhULuuuuuTJs2rZItAgAAAHAAVbNc7lvf+lauueaa3sfbt2/P/Pnz093dnZ6enkyePDlLly6tYIcAAAAAHEjVhExtbW37PJ44cWLWr19foW4AAAAAOBxVs1wOAAAAgIFLyAQAAABA0YRMAAAAABRNyAQAAABA0YRMAAAAABRNyAQAAABA0eoq3QAAwEDT2dWT5uZRlW4DAKCqCJkAAA5TfV1tFt10f0lrrlowo6T1AAD6m+VyAAAAABRNyAQAAABA0SyXAwAABoxS74nW3tGVV17eW7J6AEOZkAkAABgwSr0nmv3QAErHcjkAAAAAiiZkAgAAAKBoQiYAAAAAimZPJgAAYMiykThA6QiZAACAIctG4gClY7kcAAAAAEUTMgEAAABQNCETAAAAAEWzJxMAQ9rotw5PY4O3QwAAKJaragCGtMaGupJu+JrY9BUAgKHJcjkAAAAAiiZkAgAAAKBoQiYAAAAAiiZkAgAAAKBoVbHx98yZM9PQ0JDGxsYkycKFC3PSSSfl4YcfzpIlS9Le3p7x48dn1apVGTduXIW7BQAAAOAPVUXIlCQ333xzpkyZ0vu4UChk0aJFue6669La2ppbbrklq1evznXXXVfBLgGotNFvHZ7Ghqp5+wIAAP6fqr1Kf+SRR9LY2JjW1tYkyfnnn59Zs2YJmQCGuMaGuiy66f6S1Vu1YEbJagEAwFBWNSHTwoULUygUMn369Fx++eXZunVrjj766N7zY8eOTU9PT3bu3JmmpqY+1x03bmQ52u13zc2jKt0CDAjGytBQX1/at69S1ytHzVL/bg+E1zzU6pWjZrXXK0fNoVavHDWrvV45alb73+yBYCi+ZngzBvtYqYqQad26dWlpaUlHR0dWrFiRZcuW5bTTTitJ7e3bd6enp1CSWpXS3DwqL764q9JtQNUzVoaG5uZR6ezsKmnNUtcrR81S/m6X479hUvrXPNTqlaNmtdcrR82hVq8cNau9XjlqVvPf7IHANRj0zWAZK7W1NQec0FMVd5draWlJkjQ0NGTevHl56KGH0tLSki1btvQ+Z8eOHampqTmsWUwAAAAA9I+Kz2Tas2dPuru7M2rUqBQKhdx1112ZNm1ajjvuuLz22mt58MEH09ramttvvz1nnHFGpdsFgH7X2dUz6KdWAwAw8FU8ZNq+fXvmz5+f7u7u9PT0ZPLkyVm6dGlqa2tz/fXXZ+nSpWlvb8/48eOzatWqSrcLAP2uvq7WZucAAFS9iodMEydOzPr16/d77gMf+EA2btzYzx0BAAAAcLiqYk8mAAAAAAY2IRMAAAAARRMyAQAAAFC0iu/JBMDgNvqtw9PY4O0GAAAGO1f9AANYOQKc9o6uvPLy3pLVa2yoc2c0AAAYAoRMAANYqQOcRIgDAAC8OUImAPbR2dWT5uZRlW4DAAAYYIRMAOyjvq7W8jYAAOCwubscAAAAAEUTMgEAAABQNMvlAAAASqTUexuW+q6vAOUkZAIAACgRexsCQ5nlcgAAAAAUTcgEAAAAQNGETAAAAAAUTcgEAAAAQNGETAAAAAAUTcgEAAAAQNGETAAAAAAUTcgEAAAAQNGETAAAAAAUra7SDQAMJaPfOjyNDf70AgAAg49POgD9qLGhLotuur9k9VYtmFGyWgAAAMWwXA4AAACAogmZAAAAACiakAkAAACAolV8T6aXXnopV1xxRX7zm9+koaEh73znO7Ns2bKMHTs2U6dOzZQpU1Jb+/ss7Prrr8/UqVMr3DEAAAAAf6jiIVNNTU0uueSSnHjiiUmSlStXZvXq1fnCF76QJLn99tvzlre8pZItAgAAAHAIFQ+ZmpqaegOmJHn/+9+f2267rYIdAQNVZ1dPmptHlbRme0dXXnl5b0lrAgAADEYVD5ler6enJ7fddltmzpzZe+zCCy9Md3d3Tj755MyfPz8NDQ2HVXPcuJGlbrMiSv3BGQarq2/5z5LW+8Kn/r+Sj7/6+tL+6S11vXLUHGr1ylGz2uuVo+ZQq1eOmtVerxw1h1q9ctSs9nrlqFnt9QbCZ4GB0CNUg8E+VqoqZFq+fHlGjBiRCy64IEly3333paWlJbt3786iRYuyZs2aXHbZZYdVc/v23enpKZSj3X7T3DwqL764q9JtQNVrbh6Vzs6uktct5fgrR4/leM3V3mO11ytHzWqvV46aQ61eOWpWe71y1Bxq9cpRs9rrlaNmtder9s8CPq9A3wyWsVJbW3PACT1Vc3e5lStX5tlnn82NN97Yu9F3S0tLkmTkyJGZO3duHnrooUq2CAAAAMABVEXIdMMNN+TRRx/NmjVrepfDvfzyy3nttdeSJF1dXWlra8u0adMq2SYAAAAAB1Dx5XJPPvlk1q5dm0mTJuX8889PkkyYMCGXXHJJlixZkpqamnR1deWEE07IggULKtwtAAAAAPtT8ZDp2GOPzS9+8Yv9ntu4cWM/dwMAAADAm1EVy+UAAAAAGNiETAAAAAAUTcgEAAAAQNEqvicTMHCMfuvwNDaU7s9Ge0dXXnl5b8nqlUNnV0+am0dVug0AAICqJ2QC+qyxoS6Lbrq/ZPVWLZhRslrlUl9XO+ReMwAAwJthuRwAAAAARRMyAQAAAFA0IRMAAAAARRMyAQAAAFA0IRMAAAAARRMyAQAAAFC0uko3AAAAwP51dvWkuXlUSWu2d3TllZf3lrQmQCJkAgAAqFr1dbVZdNP9Ja25asGMktYD+B9CJgAAAN60Us+2MtMKBi4hEwAAwBBSjiV4pZxtZaYVDFxCJgAAgCGk1EvwhELA/3B3OQAAAACKZiYTUDHlmKoNAABAZQiZoEqMfuvwNDaUbkh2dnWnvm5YyeqVg6naAAAAg4eQCapEY0NdyQMXt7sFAACgv9iTCQAAAICiCZkAAAAAKJrlcgNAOTZHLvV+PaWu197RlVde3luyegAAAEB5CZkGgFJvjpyUfr+ectQDAAAABg4hE7xJpb4bHAAAUJ6VHFZKQP/wCZmqNBDeWMpxNzgAABjqyrWSAyi/qg+ZnnnmmVx11VXZuXNnmpqasnLlykyaNKnSbVFm3lgAAABgYKn6kGnp0qWZN29ezj777GzYsCFLlizJv/zLv1S6LQagcsyOAgAAql+pPwtYfgf7V9Uh0/bt2/PYY4/lq1/9apLkzDPPzPLly7Njx46MHTu2TzVqa2vK2WK/GTOqseprVnu9+rrafOGf/2/J6l391x+q+tfs96Y6a1Z7vXLUHGr1ylGz2uuVo+ZQq1eOmtVerxw1h1q9ctSs9nrlqDnU6pWjZrXXK/VngUUf++OSh1a7d71WsnqUxhIpFwkAAAnoSURBVMhRR5R0H97Orp5BkVEc7DXUFAqFQj/2clgeffTRXHnllbnzzjt7j/35n/95Vq1alfe85z0V7AwAAACA16utdAMAAAAADHxVHTK1tLTk+eefT3d3d5Kku7s7L7zwQlpaWircGQAAAACvV9Uh07hx4zJt2rRs2rQpSbJp06ZMmzatz/sxAQAAANA/qnpPpiR56qmnctVVV+WVV17J6NGjs3LlyhxzzDGVbgsAAACA16n6kAkAAACA6lfVy+UAAAAAGBiETAAAAAAUTcgEAAAAQNGETAAAAAAUTchU5Z555pmcd955Of3003Peeefl17/+daVbgop76aWX8vGPfzynn3565syZk09/+tPZsWNHkuThhx/OWWedldNPPz1//dd/ne3bt1e4W6gOX/ziFzN16tT88pe/TGKswB9qb2/P0qVL85GPfCRz5szJ5z73uSSuxeAPfe9738s555yTs88+O3PmzMk999yTxFiBlStXZubMmftcbyUHHxuDcdwImarc0qVLM2/evLS1tWXevHlZsmRJpVuCiqupqckll1yStra2bNy4MRMnTszq1atTKBSyaNGiLFmyJG1tbWltbc3q1asr3S5U3M9//vM8/PDDOfroo5PEWIH9WLVqVRobG3vfWxYsWJDEtRi8XqFQyBVXXJHrr78+GzZsyKpVq3LllVemp6fHWGHImzVrVtatW5fx48fvc/xgY2MwjhshUxXbvn17HnvssZx55plJkjPPPDOPPfZY74wNGKqamppy4okn9j5+//vfny1btuSRRx5JY2NjWltbkyTnn39+7r777kq1CVWho6Mjy5Yty9KlS1NTU5Mkxgr8gVdffTXr16/PggULesfJkUce6VoM9qO2tja7du1KkuzatStve9vb8tJLLxkrDHmtra1paWnZ59jB3kcG63tMXaUb4MC2bt2ao446KsOGDUuSDBs2LG9729uydevWjB07tsLdQXXo6enJbbfdlpkzZ2br1q29MzWSZOzYsenp6cnOnTvT1NRUwS6hcm666aacddZZmThxYu8xYwX29dxzz6WpqSlf/OIX86Mf/ShvectbsmDBghxxxBGuxeB1ampqcuONN+ZTn/pURowYkVdffTW33nqrzy1wAAcbG4VCYVCOGzOZgAFt+fLlGTFiRC644IJKtwJV56c//WkeeeSRzJs3r9KtQFXr6urKc889l3e/+9355je/mYULF2b+/PnZs2dPpVuDqtLV1ZVbb701t9xyS773ve/lS1/6Ui677DJjBehlJlMVa2lpyfPPP5/u7u4MGzYs3d3deeGFF94wBQ+GqpUrV+bZZ5/N2rVrU1tbm5aWlmzZsqX3/I4dO1JTU2NmBkPWj3/84zz99NOZNWtWkuR3v/tdLr744lx44YXGCrzO0Ucfnbq6ut4lC8cff3zGjBmTI444wrUYvM7jjz+eF154IdOnT0+STJ8+PcOHD09jY6OxAvtxsM/0hUJhUI4bM5mq2Lhx4zJt2rRs2rQpSbJp06ZMmzZtQE+dg1K54YYb8uijj2bNmjVpaGhIkhx33HF57bXX8uCDDyZJbr/99pxxxhmVbBMq6m/+5m/ygx/8IPfee2/uvffevP3tb89XvvKVXHLJJcYKvM7YsWNz4okn5j//8z+T/P5uP9u3b8+kSZNci8HrvP3tb8/vfve7PP3000mSp556Ktu2bcs73/lOYwX242Cf6Qfr5/2aQqFQqHQTHNhTTz2Vq666Kq+88kpGjx6dlStX5phjjql0W1BRTz75ZM4888xMmjQpRxxxRJJkwoQJWbNmTR566KEsXbo07e3tGT9+fFatWpUjjzyywh1DdZg5c2bWrl2bKVOmGCvwB5577rlcffXV2blzZ+rq6vLZz342M2bMcC0Gf+Db3/52/umf/ql3k/zPfOYzOfXUU40Vhrxrr70299xzT7Zt25YxY8akqakpd95550HHxmAcN0ImAAAAAIpmuRwAAAAARRMyAQAAAFA0IRMAAAAARRMyAQAAAFA0IRMAAAAARRMyAQAAAFA0IRMAwB+YOXNmfvjDH1a6jUMaKH0CAEODkAkAYBD75S9/mYsvvjgnnnhipk6dWul2AIBBTMgEADCI1dXVZfbs2VmxYkWlWwEABjkhEwDAfjz++OOZM2dOpk+fns9+9rNpb28/4HN/9KMf5eSTT84///M/50/+5E/yp3/6p7njjjt6z+/atStXXHFFPvShD+WUU07JLbfckp6ennR0dKS1tTW//OUve5+7Y8eOvO9978v27duTJN/73vdy9tlnp7W1Neeff36eeOKJw3odxxxzTObOnZtjjz32MP8LAAAcHiETAMB+fOc738mXv/zl/Pu//3t+8Ytf5Jvf/OZBn79t27bs2rUr3//+97NixYosW7YsL7/8cpJk+fLl2bVrV7773e/ma1/7WjZs2JA77rgjDQ0NOe2003LnnXfu830/+MEPZty4cfn5z3+eq6++OsuWLcuPfvSjnHfeefnUpz6Vjo6Osr52AIA3Q8gEALAfF154YY466qg0NTXllFNOyeOPP37Q59fV1eXSSy9NfX19ZsyYkREjRuSZZ55Jd3d37rrrrvzd3/1dRo4cmQkTJuSiiy7Kt7/97STJnDlzsmnTpt46GzduzJw5c5Ik//qv/5rzzjsvxx9/fIYNG5Zzzz039fX1efjhh8v3wgEA3qS6SjcAAFCNmpube/99+PDheeGFFw76/KamptTV/e+l1fDhw7Nnz5689NJL6ezszNFHH9177uijj87zzz+fJPnQhz6U9vb2/OxnP8uRRx6ZJ554IqeeemqSZMuWLVm/fn2+/vWv935tZ2fnIXsBAKgEIRMAQBmNGTMm9fX12bJlS971rnclSbZu3ZqjjjoqSVJbW5vZs2dn06ZNOfLII/Nnf/ZnGTlyZJKkpaUln/jEJ/LJT36yYv0DAPSV5XIAAGU0bNiwzJ49OzfccEN2796dzZs356tf/WrOOuus3ufMmTMn3/nOd7Jx48aceeaZvcfnzp2b22+/PT/72c9SKBSyZ8+e3Hfffdm9e3efv3+hUEh7e3s6OzuTJO3t7fZ0AgDKwkwmAIAy+9znPpfly5fn1FNPTWNjY+bOnZuPfvSjveePP/743iV5J598cu/x9773vVm+fHmWLVuWZ599NkcccUQ+8IEPpLW1tc/fe/PmzZk1a1bv4/e9730ZP3587r333tK8OACA/6emUCgUKt0EAAAAAAOb5XIAAAAAFM1yOQCAPli7dm1uvfXWNxyfPn16vvzlL1ego/91ySWX5Cc/+ckbjv/t3/5tPvGJT1SgIwBgKLJcDgAAAICiWS4HAAAAQNGETAAAAAAUTcgEAAAAQNGETAAAAAAUTcgEAAAAQNH+fzwskBHraicyAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for year in range(2016, 2021):\n",
    "    plt.figure(figsize=(20,5)) \n",
    "    sns.histplot(data=df.loc[(df['year'] == year)], x=\"h_novel_1\", bins=50)\n",
    "    plt.title(year)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is a clear shift in the distribution over years. <br>\n",
    "Let's compare 2016 and 2020 closely:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJ0AAAFCCAYAAABb4KwaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nOzdf1jUZb7/8ddnBkZFQQQHGJQy3M2oVfvd+t1qS1GsUGyL3KUfZyvxuLnZtlmRR0HSalH7tabZcU+2lqc6bK0Guchq7pW21W5tuRXZD6UllwEUxF+kwMx8/3CbjUAYnA8zw/B8XFfXxczcn7fvz9At+Lru+x7D4/F4BAAAAAAAAJjIEuwGAAAAAAAAEH4InQAAAAAAAGA6QicAAAAAAACYjtAJAAAAAAAApiN0AgAAAAAAgOkInQAAAAAAAGA6QicAAAAAAACYLiLYDQTS/v1H5HZ7gt2G3+LjB6m+/nCw2wBCHnMF8A1zBfANcwXoGvME8E24zBWLxdCQIQNP+HqfCp3cbk9YhE6SwuY+gJ7GXAF8w1wBfMNcAbrGPAF80xfmCtvrAAAAAAAAYDpCJwAAAAAAAJiuT22vAwAAAAAA8JXL1ar9+/eqtbXZ1Lp1dRa53W5Ta/Yki8WqAQMGadCgwTIMw+frCJ0AAAAAAAA6sH//XvXvH6WBA5O6FbZ0JSLCotbW3hE6eTweuVytOnSoUfv371VcXILP17K9DgAAAAAAoAOtrc0aODDG1MCptzEMQxERkYqNjVdz89FuXUvoBAAAAAAAcAJ9OXD6JsOwSOreJ+4ROgEAAAAAAMB0hE4AAAAAAAAwHaETAAAAAAAATMen1wEAAAAAEIbihkTJGmH1q4ar1aWG/U0mdYRAaW1tVURE8COfgHVQWVmpvLw8NTY2KjY2VkVFRRoxYkSbMS6XS4sXL9a2bdtkGIZmzpyp7OxsSVJ9fb3uu+8+OZ1OtbS06Pvf/77mz58fEm8iAAAAAAChxhph1b7dlX7VGJp6mknd4Gv/+79rVVHxoRYvXuJ97tFHl8hiserWW/9Ty5c/orfeekOGYdGVV07Rrbf+p6xWq/75zz0qKlqszz//TIYhXXjhOP3yl/cqOjpaknTttVM0bdo1+uMfy1RV9Q/98Y/bgp6ZBGx7XUFBgXJycrRp0ybl5OQoPz+/3ZiSkhJVVVWpvLxcL774opYvX649e/ZIklatWqWRI0eqpKREJSUl+uijj1ReXh6o9gEAAAAAAPw2adKVeuutP+vQoUOSjq9K2rLlj8rIuFIPPFAgq9WqF15YrzVr1umvf31LJSXrJUkej0c33vhTrV//Bz333O9UV1erp5/+7za1N28u15Ilj6msbGvQAycpQKFTfX29KioqlJmZKUnKzMxURUWFGhoa2ozbuHGjsrOzZbFYFBcXp/T0dJWVlUk6/hGFR44ckdvtVnNzs1paWpSYmBiI9gEAAAAAAEwxdOhQnX32udq6dbMk6e2331RsbKwSEhL01lt/1h133KUBAwZoyJA4XXddjrZsOb7gZvjwFF1wwfdls9k0ZMgQTZ9+vd5//902ta+9droSE5PUr1//gN9XRwISezmdTiUmJspqPb6X1Gq1KiEhQU6nU3FxcW3GJScnex87HA7V1NRIkm677Tbdfvvtuvjii/XVV1/p+uuv13nnndetPuLjB5lwN6HBbo8OdgtAr8BcAXzDXAF8w1wBusY8CS1RUTa/a/Tl72ldnUUREeav17nqqil6+eVi/ehH1+iPf/yDrrjiKu3dW6vW1lZlZU32jnO7PUpMTFREhEUNDQ165JEl2rHjPR050iSPx63o6Jg2/Tkcjh7p92sWi6Vb/z8Ef62Vj8rKyjRq1Cj99re/1ZEjR5Sbm6uysjJNnjy564v/pb7+sNxuTw92GRh2e7T27j0U7DaAkMdcAXzDXAF8w1wBusY8CS12e7Sampr9qhElmfI97a2HmrvdbrW2uk2ve+mll6mo6EF9+umneuONbZo1a44iIiIUGWlTaenmdlvjWlvdWrHi1/J4pGeeeV6DB8fq9df/pEcfXdKmP7fb0yP9/ru+u83/DxaL0ekCn4CETg6HQ7W1tXK5XLJarXK5XKqrq5PD4Wg3rrq6WmPGjJHUduXTc889pwcffFAWi0XR0dEaP3683n777W6FTgAAAAAAIPA41Lytfv366fLLJ6iwcL7S0s5SUlKSJOnCCy/SE088ptzcWRowIEpOZ7Xq6mp1zjnnqampSYMGDdKgQdHau7dOzz+/Nsh30bWAnOkUHx+vtLQ0lZaWSpJKS0uVlpbWZmudJE2ePFnFxcVyu91qaGjQ5s2blZGRIUkaPny4Xn/9dUlSc3Oz3nzzTX33u98NRPsAAAAAAACmmjw5U7t2fa6MjCu9z82ff79aW1t0ww3X6YorLtf8+feovn6fJOnmm3P1ySc7NXnyZbr77l/o0kvHB6t1nxkejycg+8127dqlvLw8HTx4UDExMSoqKlJqaqpyc3M1Z84cjR49Wi6XS/fff7/eeOMNSVJubq6mT58uSaqqqlJBQYH27dsnl8uliy66SP/1X//VrdPY2V4H9C3MFcA3zBXAN8wVoGvMk9Bit0ebsrrIjO9pKPXSHTU1/1BS0qmm142IsGjPnmpdf/01euWVTRo4sHecQf3t9yMkttdJ0siRI1VcXNzu+dWrV3u/tlqtKiws7PD6U045RWvWrOmx/gAAAAAAAALB7XbrxRfXacKESb0mcDoZveYgcQAAAAAAgN7uq6++0tSpk5SY6NDDD/862O30KEInAAAAAACAABkwYIC2bn2jRz9lLlQE5CBxAAAAAAAA9C2ETgAAAAAAADAdoRMAAAAAAABMR+gEAAAAAAAA03GQOAAAAAAAgI/ihkTJGmE1va6r1aWG/U2m1w0mQicAAAAAAAAfWSOs2re70q8aFosht9vT5rmhqad1ed2BA41atChf//znHtlsNg0blqK7756nIUOG6MMPP9DSpQ/q2LFjcjgcys9fpCFD4iRJhYXz9be/vaP6+n0qL39dUVFR3poHDx7Qww8X6ZNPdioiIkITJkzUzTfn+nV/3vs0pQoAAAAAAAB6lGEYysm5Sc8//7J++9sXNGzYcK1atVwej0eLFi3QL395r1544WWNHXuOnnxyufe6zMwsPfPM/3ZY84EHFuqss76nF154Wc8993/KyvqRaf0SOgEAAAAAAPQCMTGDde6553sfn3XW91RTU6OdOytks9k0duzZkqRp067V1q2bvePOO+8C76qnb/ryyyrt2vW5srN/4n0uLi7etH7ZXgcAAAAAANDLuN1u/f73L+niiy9VbW2NkpIc3tdiY2Pldrt18OABxcQMPmGNL77YLbs9Qb/61SJ99tkniouL12233aHU1JGm9MhKJwAAAAAAgF7m0UeXKipqgK655rqTruFyuVRR8aGuvHKKnn56naZMmaa8vF+a1iOhEwAAAAAAQC/yxBOPac+eKhUWPiSLxaLExCTV1Di9rzc2NsowjE5XOUlSUpJDCQlJGjv2HEnSD384XvX1+9TY2GhKn4ROAAAAAAAAvcRTT63QJ598rIceelg2m02SNGpUmo4dO6YdO96XJK1f/zuNHz+xy1qjRqVpwID+2r17lyTp/ff/ppiYwRo8uPOwylec6QQAAAAAAOAjV6tLQ1NP65G6Xdm9e5eefXaNUlJO0axZt0iSHI5kPfTQMi1YcL+WLn1Qzc3NSkpyKD9/kfe6efPu1scffyRJysm5RqmpI/XII0/IMAzNm1eghx4qVHNzi/r3768HHlgiwzBMuSdCJwAAAAAAAB817G/yu0ZEhEWtre5uX5eaOlLbt7/T4WujR4/V2rUvdvjagw8uPWHNM844U6tXr+12L75gex0AAAAAAABMR+gEAAAAAAAA0xE6AQAAAAAAwHSETgAAAAAAADAdB4kDAAAAAICQ5/F4ZLdH+1XD1eoy5SBw+IbQCQAAAAAAhDzDMLRvd6VfNYamnmZSN/AFoRMAAAAAAICPBg/uL5st0vS6zc0tOnDgaKdjDhxo1KJF+frnP/fIZrNp2LAU3X33PA0ZMkQffviBli59UMeOHZPD4VB+/iINGRKnqqp/aOnSB1Vfv09Wq1VpaWfprrvuVb9+/SVJ27e/rpUrH5fL5dKoUWmaN69A/fv3N+WeAhY6VVZWKi8vT42NjYqNjVVRUZFGjBjRZozL5dLixYu1bds2GYahmTNnKjs7W5J0zz336JNPPvGO/eSTT7RixQpNmDAhULcAAAAAAAD6OJstUssWr/KrhsViyO32tHlu7vxZkjoPnQzDUE7OTTr33PMlSStWPK5Vq5YrL2+BFi1aoHnzFmrs2LP1zDO/0ZNPLte8eQWKjIzU7bffqdNPP0Nut1sLF/6Xnn/+Of30pzPU1NSkJUse0IoVq5WScop+9atFev75Z3Xzzbl+3Z/3Pk2p4oOCggLl5ORo06ZNysnJUX5+frsxJSUlqqqqUnl5uV588UUtX75ce/bskSQtWbJEGzZs0IYNG1RUVKTBgwfrkksuCVT7AAAAAAAAQRUTM9gbOEnSWWd9TzU1Ndq5s0I2m01jx54tSZo27Vpt3bpZkuRwJOv008+QJFksFqWlnaWaGqck6a23/qwzzkhTSsop/7ruGm3Z8kfT+g1I6FRfX6+KigplZmZKkjIzM1VRUaGGhoY24zZu3Kjs7GxZLBbFxcUpPT1dZWVl7er97ne/05QpU2Sz2QLRPgAAAAAAQEhxu936/e9f0sUXX6ra2holJTm8r8XGxsrtduvgwQNtrjl27KheffUVXXzxpZKk2toaJSb++7rExCTV1dWa1mNAttc5nU4lJibKarVKkqxWqxISEuR0OhUXF9dmXHJysvexw+FQTU1Nm1rNzc0qKSnRM8880+0+4uMHndwNhCB/T+wH+grmCuAb5grgG+YK0DXmSWiJivJ/sYZZ31Mzegn0/dTVWRQR0X69jsVi+N1HRzU6+rNOZOnSIg0cGKXp03+sP/1pqwzD+Nb1hqzWf/ff2tqqhQv/S+eff4Euu+xybw8Wy7//XKvV0mkfFoulW+9frztIfPPmzUpOTlZaWlq3r62vP9xuz2RvZLdHa+/eQ8FuAwh5zBXAN8wVwDfMFaBrzJPQYrdHq6mp2a8aUZIp31Ozegn0/bjdbrW2ujt43r9soaMznSR1+Gd15IknHlNVVZWKih6V2y3Z7QlyOqu91zc2NsowpIEDo9Xa6pbL5VJh4XwNGhStO+6Y6x1ntyfq3Xf/6n1cXV2thITEE/bhdrvbvH8Wi9HpAp+AhE4Oh0O1tbVyuVyyWq1yuVyqq6uTw+FoN666ulpjxoyR1H7lkyS99NJLuuaaawLRNgAAAAAA3RY3JErWCKtfNVytLjXsbzKpI4STp55aoU8++VhLlz7uPXZo1Kg0HTt2TDt2vK+xY8/W+vW/0/jxEyUdD4oefHChLBaL8vIWyDD+vcLq+98fp0cfXaIvv6xSSsopWr/+JY0fn25arwEJneLj45WWlqbS0lJlZWWptLRUaWlpbbbWSdLkyZNVXFysSZMmqbGxUZs3b9a6deu8r9fU1Ojdd9/Vww8/HIi2AQAAAADoNmuEVft2V/pVY2jqaSZ1A7M1N7f865PmzK/bld27d+nZZ9coJeUUzZp1i6TjB4U/9NAyLVhwv5YufVDNzc1KSnIoP3+RpOOHhW/a9Aelpo7UrbfeKEkaPXqs7rrrXkVFDdQ998zTPff8Qm63W9/97ijdccdc0+4pYNvrFi5cqLy8PK1cuVIxMTEqKiqSJOXm5mrOnDkaPXq0srKytGPHDk2aNEmSNHv2bKWkpHhr/P73v9fll1+u2NjYQLUNAAAAAECf5fF4OKfrWw4cOCrpqF81IiIsPm+l+6bU1JHavv2dDl8bPXqs1q59sd3z/+//XXzCayTpkksu0yWXXNbtXnwRsNBp5MiRKi4ubvf86tWrvV9brVYVFhaesMbPfvazHukNAAAAAAC0ZxiG36u2JFZu9VW+H4sOAAAAAAAA+IjQCQAAAAAAAKYjdAIAAAAAADgBj8cT7BZCgsfjlmR0Oe6bCJ0AAAAAAAA6EBFh05EjB/t08OTxeNTa2qLGxn2y2fp369qAHSQOAAAAAADQmwwZYtf+/Xt1+HCjqXUtFovc7u5/el2wWCxWDRgwSIMGDe7WdYROAAAAAAAAHbBaIzR0qMP0unZ7tPbuPWR63VDD9joAAAAAAACYjtAJAAAAAAAApiN0AgAAAAAAgOk40wkAAAAA0OvFDYmSNcLqdx1Xq0sN+5tM6AgAoRMAAAAAoNezRli1b3el33WGpp5mQjcAJLbXAQAAAAAAoAcQOgEAAAAAAMB0hE4AAAAAAAAwHaETAAAAAAAATEfoBAAAAAAAANMROgEAAAAAAMB0hE4AAAAAAAAwHaETAAAAAAAATEfoBAAAAAAAANMROgEAAAAAAMB0hE4AAAAAAAAwHaETAAAAAAAATBew0KmyslLTp09XRkaGpk+fri+++KLdGJfLpcLCQqWnp2vixIkqLi5u8/rGjRs1ZcoUZWZmasqUKdq3b1+AugcAAAAAAEB3RATqDyooKFBOTo6ysrK0YcMG5efna+3atW3GlJSUqKqqSuXl5WpsbNS0adM0btw4DR8+XB988IGeeOIJ/fa3v5XdbtehQ4dks9kC1T4AAAAAAAHj8Xhkt0cHuw3ALwEJnerr61VRUaE1a9ZIkjIzM7Vo0SI1NDQoLi7OO27jxo3Kzs6WxWJRXFyc0tPTVVZWphkzZuiZZ57RLbfcIrvdLkmKjmbyAQAAAADCk2EY2re70q8aQ1NPM6kb4OQEJHRyOp1KTEyU1WqVJFmtViUkJMjpdLYJnZxOp5KTk72PHQ6HampqJEm7du3S8OHDdf3116upqUkTJ07Uz372MxmG4XMf8fGDTLqj4CPxBnzDXAF8w1wBfMNcAboWzHkSFWXObhgz7sGMXsKphll1zKgRKn+Xh0ofPSlg2+v85XK59Mknn2jNmjVqbm7WjBkzlJycrGnTpvlco77+sNxuTw92GRh2e7T27j0U7DaAkMdcAXzDXAF8w1wBuhbMeWK3R6upqdnvOlGS3/dgRi9RUtjUCKVezPj+miFcfqZYLEanC3wCcpC4w+FQbW2tXC6XpOMBUl1dnRwOR7tx1dXV3sdOp1NJSUmSpOTkZE2ePFk2m02DBg3ShAkT9Pe//z0Q7QMAAAAAAKCbAhI6xcfHKy0tTaWlpZKk0tJSpaWltdlaJ0mTJ09WcXGx3G63GhoatHnzZmVkZEg6fg7U9u3b5fF41NLSorfeektnnHFGINoHAAAAAABANwVse93ChQuVl5enlStXKiYmRkVFRZKk3NxczZkzR6NHj1ZWVpZ27NihSZMmSZJmz56tlJQUSdJVV12lDz/8UFdeeaUsFosuvvhiXXvttYFqHwAAAAAAAN0QsNBp5MiRKi4ubvf86tWrvV9brVYVFhZ2eL3FYtF9992n++67r8d6BAAAAAAAgDkCsr0OAAAAAAAAfQuhEwAAAAAAAExH6AQAAAAAAADTBexMJwAAAAAAQp3H45HdHh3sNoCwQOgEAAAAAMC/GIahfbsr/aoxNPU0k7oBeje21wEAAAAAAMB0hE4AAAAAAAAwHaETAAAAAAAATEfoBAAAAAAAANMROgEAAAAAAMB0hE4AAAAAAAAwHaETAAAAAAAATEfoBAAAAAAAANMROgEAAAAAAMB0hE4AAAAAAAAwHaETAAAAAAAATBcR7AYAAAAAAH3b4MH9ZbNFBrsNACYjdAIAAAAABJXNFqlli1f5VWPu/FkmdQPALGyvAwAAAAAAgOkInQAAAAAAAGA6QicAAAAAAACYzufQ6aGHHtLHH3/ck70AAAAAAAAgTPh8kHhra6tuvfVWxcXFaerUqZo6daqSkpJ6sjcAAAAAAAD0Uj6vdFqwYIG2bdumu+66Szt37tQVV1yhn/70p1q/fr2OHDnS5fWVlZWaPn26MjIyNH36dH3xxRftxrhcLhUWFio9PV0TJ05UcXGx97Xly5dr3LhxysrKUlZWlgoLC31tHQAAAAAAAAHm80onSbJarbr88st1+eWX67PPPtNdd92lvLw8FRYW6sorr9ScOXOUmJjY4bUFBQXKyclRVlaWNmzYoPz8fK1du7bNmJKSElVVVam8vFyNjY2aNm2axo0bp+HDh0uSpk2bpnvvvfckbxUAAAAA8E2DB/eXzRbpV43m5hYdOHDUpI4AhJNuhU6HDx/WH/7wB73yyiv69NNPNWnSJBUUFCg5OVlPP/20ZsyYoZKSknbX1dfXq6KiQmvWrJEkZWZmatGiRWpoaFBcXJx33MaNG5WdnS2LxaK4uDilp6errKxMM2bM8PM2AQAAAADfZrNFatniVX7VmDt/liRCJwDt+Rw6zZkzR9u2bdMFF1ygn/zkJ0pPT5fNZvO+ft999+m8887r8Fqn06nExERZrVZJx1dMJSQkyOl0tgmdnE6nkpOTvY8dDodqamq8j1999VVt375ddrtdt99+u8455xzf71RSfPygbo0PZXZ7dLBbAHoF5grgG+YK4BvmCsJR1MB+ftf45tw42XliRh9RUbauBwWoDjV6po4ZNULl7/JQ6aMn+Rw6jR07VgsWLJDdbu/wdYvFoj//+c+mNfZtP/7xjzVr1ixFRkbqjTfe0G233aaNGzdqyJAhPteorz8st9vTYz0Git0erb17DwW7DSDkMVcA3zBXAN8wVxCO7PZoNR055nedr+fGyc4Ts/poamr2u0aUCXWoEbq9REkh8Xd5uPxMsViMThf4+HyQ+DvvvNNh4PTzn//c+/WAAQM6vNbhcKi2tlYul0vS8QPD6+rq5HA42o2rrq72PnY6nd5PyLPb7YqMPL7X+Ac/+IEcDoc+++wzX9sHAAAAAABAAPkcOr399tsdPv+Xv/yly2vj4+OVlpam0tJSSVJpaanS0tLabK2TpMmTJ6u4uFhut1sNDQ3avHmzMjIyJEm1tbXecR9//LH++c9/6rTTTvO1fQAAAAAAAARQl9vrHn/8cUlSS0uL9+uvffnll23OYOrMwoULlZeXp5UrVyomJkZFRUWSpNzcXM2ZM0ejR49WVlaWduzYoUmTJkmSZs+erZSUFEnSI488oo8++kgWi0WRkZFasmTJCbf6AQAAAAAAILi6DJ2+Psjb4/G0OdRbOr4d7vbbb/fpDxo5cqSKi4vbPb969Wrv11arVYWFhR1e/3VIBQAAAAAAgNDXZej00EMPSZLOOeccXXfddT3eEAAAAACga4MH95fNFhnsNtTa2mrKp9cBCD+dhk579uzR8OHDJUnjxo3Tl19+2eG4r7fAAQAAAAACw2aL1LLFq/yqMXf+LL/7iIiI8PYRNbDfSX0KnRl9SFJUlM2UOgDM0WnoNGXKFL333nuSpIkTJ8owDHk8njZjDMPQxx9/3HMdAgAAAADggze3vet3jQnZw0zoBKHK4/H4vRrP1epSw/4mkzoKb52GTl8HTpK0c+fOHm8GAAAAAMJdqGyLA/oiwzC0b3elXzWGpp5mUjfhr8sznQAAAAAA5jFjW5xk3pY0AOgpnYZOOTk5MgyjyyLr1q0zrSEAAAAAAAD0fp2GTtnZ2YHqAwAAAACAsMGh5kAXodPVV18dqD4AAAAAAAgJZgRG/h5qzoHmCAedhk7r16/XtGnTJEm/+93vTjju2muvNbcrAAAAAACChMAIMEenodOrr77qDZ02bNjQ4RjDMAidAAAAAAAA0EanodPq1au9Xz/77LM93gwAAAAAAADCQ6eh07cdPHhQf/rTn1RXV6eEhARddtlliomJ6aneAAAAAAAA0EtZfB345ptvavz48Xr22Wf1wQcf6LnnntP48eP15ptv9mR/AAAAAAAA6IV8Xum0aNEi3X///bryyiu9z/3hD39QYWGhysrKeqQ5AAAAAAglgwf3l80WGew2AKBX8Dl0qqurU0ZGRpvnJk6cqAULFpjeFAAAAACEIpstUssWr/Krxtz5s0zqBgBCm8/b66ZNm6Z169a1ee7555/3frodAAAAAAAA8LVOVzrl5OTIMAxJktvt1vPPP6/f/OY3SkxMVG1trerr6zV27NiANAoAAAAAAIDeo9PQKTs7u83j6667rkebAQAAAAAAQHjoNHS6+uqrA9UHAAAAAABAyPN4PLLbo/2u0Rf4fJC4JO3bt09///vftX///jZv0LXXXmt6YwAAAAAAAKHGMAzt213pV42hqaeZ1E1o8zl02rx5s+6++26deuqp+vzzz/Wd73xHn332mc4991xCJwAAAAAAALThc+j02GOP6cEHH9QVV1yhCy64QOvXr9dLL72kzz//vCf7AwAAAAD0EI/Ho7nzZ/ldAwA64nPoVF1drSuuuKLNc1dffbV+8IMf6N577+3y+srKSuXl5amxsVGxsbEqKirSiBEj2oxxuVxavHixtm3bJsMwNHPmzHaHme/evVtXX321cnJyfPpzAQAAAAAdMwxDW4pf8avGhOypJnUDINxYfB0YHx+vffv2SZKGDRum9957T1VVVXK73T5dX1BQoJycHG3atEk5OTnKz89vN6akpERVVVUqLy/Xiy++qOXLl2vPnj3e110ulwoKCpSenu5r2wAAAAAAAAgCn0On7Oxsvfvuu5Kkn/70p7rpppuUlZWln/zkJ11eW19fr4qKCmVmZkqSMjMzVVFRoYaGhjbjNm7cqOzsbFksFsXFxSk9PV1lZWXe1//7v/9bl112WbsVUgAAAAAAAAgtPm+vmzlzpvfradOm6cILL9RXX32lkSNHdnmt0+lUYmKirFarJMlqtSohIUFOp1NxcXFtxiUnJ3sfOxwO1dTUSJJ27typ7du3a+3atVq5cqWvbQMAAAAAACAIfA6dpOPb295//33V1dUpISFBZ599dk/11UZLS4sWLFighx56yBtcnYz4+EEmdhVcdnt0sFsAegXmCuAb5grgG+YKJClqYL+QqGFWncjIk/83Vkd9nGxPoXIvZtUx5X2NsoVNDbPqhFONvvAzxefQaefOnZo9e7aOHTumpKQk1dTUqF+/flqxYoXOOOOMTq91OE68o/8AAB9tSURBVByqra2Vy+WS1WqVy+VSXV2dHA5Hu3HV1dUaM2aMpH+vfNq7d6+qqqq8q60OHjwoj8ejw4cPa9GiRT7fbH39Ybndvf+TFez2aO3deyjYbQAhj7kC+Ia5AviGuQLp+P8HTUeO+V3HjBpm1WlpcZnWR9TAfifdU6jci1l1THlfm5r9uj4qRGqEUi+hVCMcfqZYLEanC3x8Dp3mzZun66+/XjfffLMMw5DH49EzzzyjefPm6eWXX+702vj4eKWlpam0tFRZWVkqLS1VWlpam611kjR58mQVFxdr0qRJamxs1ObNm7Vu3TolJyfr7bff9o5bvny5mpqa+PQ6AAAAAACAEOVz6PTFF1/oP/7jP2QYhqTjH6150003afny5T5dv3DhQuXl5WnlypWKiYlRUVGRJCk3N1dz5szR6NGjlZWVpR07dmjSpEmSpNmzZyslJaW79wQAAAAA7Qwe3F82W2Sw2wCAPsPn0OmHP/yhXnvtNU2cONH73NatW3XZZZf5dP3IkSNVXFzc7vnVq1d7v7ZarSosLOyy1u233+7TnwkAAAAAX7PZIrVs8Sq/asydP8ukbgAg/HUaOt19993elU0ul0t33nmnvve973nPdPrwww81YcKEgDQKAAAAAACA3qPT0OnUU09t8/j000/3fv2d73xHF198cc90BQAAAAAAgF6t09Dp5z//eaD6AAAAAAD0UR6Ph62LQBjy+UwnSXrrrbe0YcMG1dXVKSEhQVOnTtW4ceN6qjcAAAAAQB9gGIa2FL/iV40J2VNN6gaAWSy+DiwuLtadd94pu92uiRMnKiEhQXPnztX//d//9WR/AAAAAAAA6IV8Xun0m9/8RmvWrNEZZ5zhfe6KK67QnDlzdN111/VIcwAAAAAAAOidfF7p1NjYqJEjR7Z5LjU1VQcOHDC9KQAAAAAAAPRuPodO5557rn71q1/pq6++kiQ1NTVpyZIlOuecc3qsOQAAAAAAAPROPm+vKyws1F133aXzzz9fgwcP1oEDB3TOOefo4Ycf7sn+AAAAAAAA0Av5FDp5PB4dO3ZMa9as0b59+7yfXpeUlNTT/QEAAAAAAKAX8ml7nWEYmjJliiwWi5KSkjRmzBgCJwAAAAAAAJyQz2c6paWlqbKysid7AQAAAAAAQJjw+UynCy+8ULm5ubr66quVlJQkwzC8r1177bU90hwAAAAAILR5PB7NnT8r2G0ACEE+h05/+9vfNGzYMP3lL39p87xhGIROAAAAANBHGYahLcWvSJIiI61qaXF1u8aE7KlmtwUgBHQZOn311Vd68sknNXDgQJ155pmaNWuWbDZbIHoDAAAAAABAL9XlmU7333+/tm7dqtTUVJWXl6uoqCgQfQEAAAAAAKAX63Kl07Zt2/Tyyy8rISFBN954o66//notWLAgEL0BAAAAANBnRUX5v8vIjBrAyeoydGpqalJCQoIkyeFw6PDhwz3eFAAAAAAAfd2b29716/oJ2cNMqQGcrC5DJ5fLpbfeeksej0eS1Nra2uaxJI0bN67nOgQAAAAQFIMH95fNFul3nebmFh04cNSEjgAAvUmXoVN8fLzmzZvnfRwbG9vmsWEY2rJlS890BwAAACBobLZILVu8yu86c+fPkkToBAB9TZeh02uvvRaIPgAAAAAAABBGuvz0OgAAAAAAAKC7CJ0AAAAAAABgui6315mlsrJSeXl5amxsVGxsrIqKijRixIg2Y1wulxYvXqxt27bJMAzNnDlT2dnZkqSXXnpJzzzzjCwWi9xut7Kzs3XTTTcFqn0AAAAAQWTWoeb+8ng8/zqjyr8aANAXBCx0KigoUE5OjrKysrRhwwbl5+dr7dq1bcaUlJSoqqpK5eXlamxs1LRp0zRu3DgNHz5cGRkZ+tGPfiTDMHT48GFNmTJFF154oc4444xA3QIAAACAIDHjUHN/wyLpXx+kVPyKXzUmZE/1uw8A6A0Csr2uvr5eFRUVyszMlCRlZmaqoqJCDQ0NbcZt3LhR2dnZslgsiouLU3p6usrKyiRJgwYNkmEYkqSjR4+qpaXF+xgAAAAAAAChJSChk9PpVGJioqxWqyTJarUqISFBTqez3bjk5GTvY4fDoZqaGu/jLVu26KqrrtLll1+uGTNmaNSoUYFoHwAAAAAAAN0UsO11ZpgwYYImTJig6upqzZ49W5deeqlSU1N9vj4+flAPdhdYdnt0sFsAegXmCuAb5grgm744V6IG9jOljhnvnRm9mFEjMtIaEn2YVceM+/lmjZOtZ3Yfwa4TTjWiomx+1zCrTjjV6As/UwISOjkcDtXW1srlcslqtcrlcqmurk4Oh6PduOrqao0ZM0ZS+5VPX0tOTtbo0aP1pz/9qVuhU339Ybndvf/QPrs9Wnv3Hgp2G0DIY64AvmGuAL7pi3PFbo9W05FjptTy970zqxczarS0uEKiD7PqmHE/X9eIjLSedD0z+wiFOuFUo6mp2e8aUSbUCbca4fAzxWIxOl3gE5DtdfHx8UpLS1NpaakkqbS0VGlpaYqLi2szbvLkySouLpbb7VZDQ4M2b96sjIwMSdKuXbu84xoaGvT222/r9NNPD0T7AAAAAAAA6KaAba9buHCh8vLytHLlSsXExKioqEiSlJubqzlz5mj06NHKysrSjh07NGnSJEnS7NmzlZKSIkl68cUX9cYbbygiIkIej0c33HCDLr744kC1DwAAAAAAgG4IWOg0cuRIFRcXt3t+9erV3q+tVqsKCws7vH7evHk91hsAAAAAAADM1asOEgcAAADQ+7S2tvaJA3MBAG0ROgEAAADoUREREVq2eJVfNebOn2VSN+HD4/HwvgAIaYROAAAAABBAZoZFW4pf8ev6CdlTTekDADpC6AQAAAAAAWQYht9hkURgBCD0WYLdAAAAAAAAAMIPoRMAAAAAAABMR+gEAAAAAAAA0xE6AQAAAAAAwHSETgAAAAAAADAdn14HAAAAAABOKCrKFuwW0EsROgEAAAAAgBN6c9u7fteYkD3MhE7Q27C9DgAAAAAAAKYjdAIAAAAAAIDp2F4HAAAAhJjBg/vLZov0q0Zzc4sOHDhqUkcAAHQfoRMAAAAQYmy2SC1bvMqvGnPnz5JE6AQACB621wEAAAAAAMB0hE4AAAAAAAAwHdvrAAAAAPQJHo/nX9sOAQCBQOgEAAAAmMiMQ8DRMwzD0JbiV/yqMSF7qkndAED4I3QCAAAATGTeIeD+aW1tld0e7XcdAABOFqETAAAAEIYiIiJCIvwCAPRdHCQOAAAAAAAA0xE6AQAAAAAAwHQBC50qKys1ffp0ZWRkaPr06friiy/ajXG5XCosLFR6eromTpyo4uJi72srVqzQVVddpalTp+pHP/qRtm3bFqjWAQAAAAAA0E0BO9OpoKBAOTk5ysrK0oYNG5Sfn6+1a9e2GVNSUqKqqiqVl5ersbFR06ZN07hx4zR8+HCNGTNGt9xyiwYMGKCdO3fqhhtu0Pbt29W/f/9A3QIAAAAAAAB8FJCVTvX19aqoqFBmZqYkKTMzUxUVFWpoaGgzbuPGjcrOzpbFYlFcXJzS09NVVlYmSbrkkks0YMAASdKoUaPk8XjU2NgYiPYBAAAAAADQTQEJnZxOpxITE2W1WiVJVqtVCQkJcjqd7cYlJyd7HzscDtXU1LSrt379ep1yyilKSkrq2cYBAAAAAABwUgK2vc4sf/nLX/T444/r6aef7va18fGDeqCj4LDbo4PdAtArMFcA3zBXgONaW1sVEXHiX5F9nStRA/v53Us41TCrjhk1IiOtYVPDrDpm1zjZeqFyL2bVoUZ7UVE2anxDX/j9KyChk8PhUG1trVwul6xWq1wul+rq6uRwONqNq66u1pgxYyS1X/n03nvv6e6779bKlSuVmpra7T7q6w/L7fb4dzMhwG6P1t69h4LdBhDymCuAb5grwL/Z7dFatnhVh69FDeynpiPHuqwxd/4sn8Z1JZxqmFXHjBotLa6wqWFWHTNrREZaT7peqNyLWXWo0V5TU7Nf10eFWY1w+P3LYjE6XeATkO118fHxSktLU2lpqSSptLRUaWlpiouLazNu8uTJKi4ultvtVkNDgzZv3qyMjAxJ0t///nfdeeed+vWvf62zzjorEG0DAAAAAACTREXZ/PoPvU/AttctXLhQeXl5WrlypWJiYlRUVCRJys3N1Zw5czR69GhlZWVpx44dmjRpkiRp9uzZSklJkSQVFhbq6NGjys/P99ZcsmSJRo0aFahbAAAAAAAAJ+nNbe/6df2E7GEmdYJACVjoNHLkSBUXF7d7fvXq1d6vrVarCgsLO7z+pZde6rHeAAAAAAAAYK5ed5A4AAAAgMDweDyaO3+WKXVCpRcAMAPb/XxD6AQAAACgQ4ZhaEvxK37XmZA9NSR6MaMPAJDYKuirgBwkDgAAAAAAgL6FlU4AAAAAehRb4wCgbyJ0AgAAANCj2BoHAH0ToRMAAAB6vcGD+8tmiwx2GwAA4BsInQAAANDr2WyRWrZ4ld912AIGAIB5OEgcAAAAAAAApiN0AgAAAAAAgOkInQAAAAAAAGA6QicAAAAAAACYjtAJAAAAAAAApiN0AgAAAAAAgOkigt0AAAAA+rbBg/vLZosMdhsAAMBkhE4AAAAIKpstUssWr/Krxtz5s0zqBgAAmIXtdQAAAAAAADAdK50AAAACyKytZM3NLTpw4KgJHSFceTweVoABAIKK0AkAACCAzNhKJn29nSz4oRPnMYUuwzC0pfgVv2pMyJ5qUjcAgL6I0AkAAAAnjfOYAADAiRA6AQAA9FGsUgIAAD2J0AkAAKCPYpUSAADoSXx6HQAAAAAAAEzHSicAAAAgxPDJcwCAcBCw0KmyslJ5eXlqbGxUbGysioqKNGLEiDZjXC6XFi9erG3btskwDM2cOVPZ2dmSpO3bt+uRRx7Rp59+qhtvvFH33ntvoFoHAAAAAopPngMAhIOAba8rKChQTk6ONm3apJycHOXn57cbU1JSoqqqKpWXl+vFF1/U8uXLtWfPHklSSkqKFi9erFtvvTVQLQMAAAAAAOAkBSR0qq+vV0VFhTIzMyVJmZmZqqioUENDQ5txGzduVHZ2tiwWi+Li4pSenq6ysjJJ0qmnnqozzzxTERHsCAQAAAAAAAh1AUlwnE6nEhMTZbVaJUlWq1UJCQlyOp2Ki4trMy45Odn72OFwqKamxrQ+4uMHmVYr2Oz26GC3APQKzBXAN8yVwIoa2M+UOmZ838zoJZxqdFXH1z/DjF4iI61hU8OsOtTomTpm1zjZeqFyL2bVoUbP1ImKsoVEDTPupS/8/tWnlg3V1x+W2+0Jdht+s9ujtXfvoWC3AYQ85grgG+ZKYNnt0Wo6csyUWv5+38zqJZxqdFYnamA/n/4Mj8ej2+682e8+WlpcYVPDrDrU6Jk6ZtaIjLSedL1QuRez6lCjZ+o0NTX7dX2USTXMuJdw+P3LYjE6XeATkNDJ4XCotrZWLpdLVqtVLpdLdXV1cjgc7cZVV1drzJgxktqvfAIAAABCHYeAA0BoM2OlE3wTkNApPj5eaWlpKi0tVVZWlkpLS5WWltZma50kTZ48WcXFxZo0aZIaGxu1efNmrVu3LhAtAgAA9Cqtra19Ylk+AADfZEZg9Oa2d/26fkL2ML976CsCtr1u4cKFysvL08qVKxUTE6OioiJJUm5urubMmaPRo0crKytLO3bs0KRJkyRJs2fPVkpKiiTpnXfe0S9/+UsdPnxYHo9Hr776qh544AFdcsklgboFAAB6jcGD+8tmi/R5fEfhRXNziw4cOGpmW0HV3fekI6H0nkRERGjZ4lV+1Zg7f5ZJ3QAAEBgERr1LwEKnkSNHqri4uN3zq1ev9n5ttVpVWFjY4fXnn3++Xn/99R7rDwCAcGKzRfocSJzonJrjgURoBCxm6M57ciLh9p4AAAD0pD51kDgAAIA/2NIGAADgO0InAAAAH7GlDQAAwHeETgAAIOSZcR4TAAAAAovQCQAAhDzzzmMCAABAoBA6AQAQYljVAwAAgHBA6AQAQIgJlVU9Zh2a3drSqohIfuVA7+DxeFgVBwCASfgNEAAAdMiMQ7Ol4wFYKIRogC8Mw9CW4lc6fC0y0qqWFleXNSZkTzW7LQAAeiVCJwAAAPR6rFACACD0EDoBAMKCGecgNTe36MCBoyZ1BPQsM0IWj8djUjfB19kKpe5glRIAAOYhdAIAhAUzzkH6Rd4Mv88wIrhCoJgRshCwAACAnkToBADAv5hxhtHxlSeETug7WHEFAABOhNAJAAATmfWJb0BvwYorAABwIoROABBgnD3UlhnvRygxb7UUAAAA0LsROgFAgJlx9pAZW7hCJfwy4/2QCGqA3oxPngMAIDwROgGAj8JtRU6ohF8AwBY9AADCE6FTL2TGeSHhtDUH4c+fsOfrucKKHAAAAAAILEKnXohPV0Jfc7JhT9TAfmo6ckxS+P0/z2HVAAAAAEIdoROAEwqVM3/MEG4hTagcVh1u7ysAAAAA8xA6AWHIzLOHwmVVXaiENOGG9xXo3TjAGwAA9CRCJyDEmBUYcfYQAKArHOANAAB6EqETwoIZQU1rS6siIv2bEmbUkMxaXRQa2H4FIFSYsarHrBoAAAB9AaETTloonfdj1ke/h0qNcML2K6D7wi3YCJX7MWtVDyuDAAAAfEPohJNmRtDzi7wZrIIBgG8Jty1P4XY/AAAA8E3AQqfKykrl5eWpsbFRsbGxKioq0ogRI9qMcblcWrx4sbZt2ybDMDRz5kxlZ2d3+Rp6LzNWwUishAEQOqtpQkW4vR/hdOB1ON0LAABAZwIWOhUUFCgnJ0dZWVnasGGD8vPztXbt2jZjSkpKVFVVpfLycjU2NmratGkaN26chg8f3ulrAACwmqat7rwfkZFWtbS42j0fSu9HOH1/zbgXKXTuBwAA4EQCEjrV19eroqJCa9askSRlZmZq0aJFamhoUFxcnHfcxo0blZ2dLYvFori4OKWnp6usrEwzZszo9DVfWSyG6fcWLDGD/duSZtbhzv72YVYNs+pQw/w6Zq22OJk+BkTZFBFh8z4Olfc1mO/Jt4XK99esVR/9owb4XSMU3hPJnHvxtUZEpFXWDkInyby/SwJ5P32lhll1qOF7nc7mSk/0Ek41zKpDjZ6pY2aN7syTnuwjFOpQo2fqhFONcMgouroHwxOAtfMffvih7r33Xr366qve56688kotXbpUZ511lve5KVOm6IEHHtCYMWMkSatXr1Ztba3mz5/f6WsAAAAAAAAILZZgNwAAAAAAAIDwE5DQyeFwqLa2Vi7X8WWWLpdLdXV1cjgc7cZVV1d7HzudTiUlJXX5GgAAAAAAAEJLQEKn+Ph4paWlqbS0VJJUWlqqtLS0Nuc5SdLkyZNVXFwst9uthoYGbd68WRkZGV2+BgAAAAAAgNASkDOdJGnXrl3Ky8vTwYMHFRMTo6KiIqWmpio3N1dz5szR6NGj5XK5dP/99+uNN96QJOXm5mr69OmS1OlrAAAAAAAACC0BC50AAAAAAADQd3CQOAAAAAAAAExH6AQAAAAAAADTEToBAAAAAADAdIROAAAAAAAAMB2hUy9SWVmp6dOnKyMjQ9OnT9cXX3wR7JaAkLB//37l5uYqIyNDU6ZM0c9//nM1NDRIkt5//31NnTpVGRkZuuWWW1RfXx/kboHge+KJJzRq1Ch9+umnkpgnwLcdO3ZMBQUFmjRpkqZMmaIFCxZI4ncx4Nu2bt2qadOmKSsrS1OmTFF5ebkk5gpQVFSk8ePHt/l9S+p8boTrvCF06kUKCgqUk5OjTZs2KScnR/n5+cFuCQgJhmFoxowZ2rRpk0pKSpSSkqJly5bJ4/Ho7rvvVn5+vjZt2qTzzz9fy5YtC3a7QFB99NFHev/995WcnCxJzBOgA0uXLlW/fv28P1fuuOMOSfwuBnyTx+PRPffcoyVLlmjDhg1aunSp7r33XrndbuYK+rwJEyZo3bp1GjZsWJvnO5sb4TpvCJ16ifr6elVUVCgzM1OSlJmZqYqKCu9qDqAvi42N1UUXXeR9fPbZZ6u6uloffPCB+vXrp/PPP1+S9OMf/1hlZWXBahMIuubmZt1///0qKCiQYRiSxDwBvuXIkSNav3697rjjDu88GTp0KL+LAR2wWCw6dOiQJOnQoUNKSEjQ/v37mSvo884//3w5HI42z3X2cyScf8ZEBLsB+MbpdCoxMVFWq1WSZLValZCQIKfTqbi4uCB3B4QOt9ut559/XuPHj5fT6fSu5pCkuLg4ud1uNTY2KjY2NohdAsHx+OOPa+rUqUpJSfE+xzwB2vryyy8VGxurJ554Qm+//bYGDhyoO+64Q/379+d3MeAbDMPQY489pttuu01RUVE6cuSInnrqKf7dApxAZ3PD4/GE7bxhpROAsLJo0SJFRUXphhtuCHYrQEh577339MEHHygnJyfYrQAhrbW1VV9++aXOPPNMvfzyy5o7d65uv/12NTU1Bbs1IKS0trbqqaee0sqVK7V161Y9+eSTuvPOO5krANpgpVMv4XA4VFtbK5fLJavVKpfLpbq6unZL9oC+rKioSP/4xz+0atUqWSwWORwOVVdXe19vaGiQYRis3kCf9Ne//lW7d+/WhAkTJEk1NTW69dZbdeONNzJPgG9ITk5WRESEd4vD2LFjNWTIEPXv35/fxYBv+Pjjj1VXV6fzzjtPknTeeedpwIAB6tevH3MF6EBn/6b3eDxhO29Y6dRLxMfHKy0tTaWlpZKk0tJSpaWl9fqldoBZHn30UX344YdasWKFbDabJOl73/uejh49qnfeeUeS9MILL+iKK64IZptA0MycOVPbt2/Xa6+9ptdee01JSUn6n//5H82YMYN5AnxDXFycLrroIr3xxhuSjn+aUH19vUaMGMHvYsA3JCUlqaamRrt375Yk7dq1S/v27dOpp57KXAE60Nm/6cP53/uGx+PxBLsJ+GbXrl3Ky8vTwYMHFRMTo6KiIqWmpga7LSDoPvvsM2VmZmrEiBHq37+/JGn48OFasWKF/va3v6mgoEDHjh3TsGHDtHTpUg0dOjTIHQPBN378eK1atUqnn3468wT4li+//FLz5s1TY2OjIiIi9Itf/EI//OEP+V0M+JZXXnlFq1ev9h66P2fOHKWnpzNX0OctXrxY5eXl2rdvn4YMGaLY2Fi9+uqrnc6NcJ03hE4AAAAAAAAwHdvrAAAAAAAAYDpCJwAAAAAAAJiO0AkAAAAAAACmI3QCAAAAAACA6QidAAAAAAAAYDpCJwAAAAAAAJiO0AkA/n/7dhMS5RoFcPx/07EMF1bGkLQMWqWgExjElGHgoreoECNoUURZbYSoheRmBjdCtAikQnRRiyimLy03RuGilCgSCq1NtMgo+iBGDLWcdu/96F6/GrKb/9/q5XnOM8857+rlMEeSJrFp0ybu3bs312lM6f+SpyRJmj9sOkmSJM0TV69eZceOHZSVlRGPx2lububLly9znZYkSfpN2XSSJEmaJz5//kxDQwO9vb1cvnyZ3t5e2tra5jotSZL0m7LpJEmSNIWBgQGCIKC8vJz6+npGR0f/M7avr494PE5bWxvr1q1j/fr1pFKpcD+dTnP8+HEqKiqorKykpaWFiYkJxsbGiMViPH/+PIz98OEDJSUlvH//HoA7d+6wbds2YrEYu3btYnBwcEZ17N69m1gsRl5eHtFolCAIePTo0QzfhiRJ0vTYdJIkSZpCV1cXra2t3L59m2fPnnHlypVJ49+9e0c6naanp4empiYSiQSfPn0CIJlMkk6n6e7u5vz581y/fp1UKkVeXh6bN2/m5s2bf7t37dq1LFu2jKdPn9LQ0EAikaCvr4/a2loOHz7M2NjYrOt68OABq1atmvV5SZKkydh0kiRJmsKePXuIRqMUFhZSWVnJwMDApPG5ubkcOXKESCTChg0bWLx4MS9evODr16/cunWLo0ePUlBQwMqVK9m7dy83btwAIAgCOjs7w9/p6OggCAIALl26RG1tLaWlpeTk5LB9+3YikQiPHz+eVU2pVIonT56wb9++WZ2XJEmaSu5cJyBJkvSrW758eficn5/P27dvJ40vLCwkN/fPz6z8/HxGRkb4+PEj4+PjFBcXh3vFxcW8efMGgIqKCkZHR+nv76eoqIjBwUGqqqoAGBoa4tq1a1y4cCE8Oz4+PmUu/6a7u5uTJ0/S3t7O0qVLZ3xekiRpOmw6SZIk/SRLliwhEokwNDQUjrW9fv2aaDQKwIIFC6iurqazs5OioiI2btxIQUEBACtWrKCuro5Dhw79UA49PT2cOHGCc+fOsXr16h8rSJIkaRKO10mSJP0kOTk5VFdXc+rUKYaHh3n16hXt7e1s3bo1jAmCgK6uLjo6OtiyZUu4XlNTw8WLF+nv7yeTyTAyMsLdu3cZHh6e9v3379/n2LFjnD59mpKSkqzWJkmS9E/+00mSJOknamxsJJlMUlVVxcKFC6mpqWHnzp3hfmlpaTjCF4/Hw/U1a9aQTCZJJBK8fPmSRYsWUVZWRiwWm/bdLS0tpNNpDhw4EK6Vl5fT2tqaneIkSZL+4o9MJpOZ6yQkSZIkSZL0e3G8TpIkSZIkSVnneJ0kSdIMnTlzhrNnz363/iuMqu3fv5+HDx9+t37w4EHq6urmICNJkjRfOV4nSZIkSZKkrHO8TpIkSZIkSVln00mSJEmSJElZZ9NJkiRJkiRJWWfTSZIkSZIkSVln00mSJEmSJElZ9w0zP6nuhWKj7wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for n in [1,2,3]:\n",
    "    plt.figure(figsize=(20,5))\n",
    "    sns.histplot(data=df.loc[((df['year'] == 2020)) | (df['year'] == 2016)],\n",
    "                 x=\"h_novel_{}\".format(str(n)), hue=\"year\", stat=\"probability\", common_norm=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's check the distribution of heading_article_len_ratio in 2016 and 2020"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fc91c8a0640>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,5))\n",
    "sns.histplot(data=df.loc[((df['year'] == 2020)) | (df['year'] == 2016)],\n",
    "             x=\"heading_article_len_ratio\".format(str(n)), hue=\"year\", stat=\"probability\", common_norm=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fc91cb736d0>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,8))\n",
    "sns.ecdfplot(data=df[['h_novel_1']], y=\"h_novel_1\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10015.json     54.5\n",
       "1002.json      57.1\n",
       "10027.json     58.5\n",
       "10029.json     63.6\n",
       "10030.json     59.1\n",
       "10031.json     51.7\n",
       "10032.json     62.7\n",
       "10038.json     64.5\n",
       "10040.json    100.0\n",
       "10054.json     57.1\n",
       "10064.json     53.8\n",
       "1009.json      52.6\n",
       "10091.json     62.9\n",
       "10092.json     65.8\n",
       "10097.json     64.0\n",
       "10126.json     85.7\n",
       "10139.json     54.5\n",
       "10147.json     72.1\n",
       "10148.json     63.3\n",
       "10152.json     56.0\n",
       "10153.json     65.5\n",
       "10165.json     56.2\n",
       "10174.json     52.8\n",
       "10177.json     59.3\n",
       "10188.json     51.9\n",
       "10191.json     63.6\n",
       "10193.json     51.9\n",
       "10200.json     62.5\n",
       "10210.json     62.1\n",
       "10211.json     53.8\n",
       "10213.json     66.7\n",
       "10215.json     65.2\n",
       "10216.json     65.2\n",
       "10219.json     56.9\n",
       "10221.json     72.0\n",
       "10223.json     66.0\n",
       "1023.json      52.0\n",
       "10245.json     59.3\n",
       "10254.json     53.8\n",
       "10255.json     63.6\n",
       "Name: h_novel_1, dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df['h_novel_1'] > 50]['h_novel_1'][:40]"
   ]
  }
 ],
 "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}