Project_2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Finding the Best Time to Create Posts on Hacker News\n",
    "---\n",
    "[Hacker News](https://news.ycombinator.com/news) is a site started by the startup incubator [Y Combinator](https://www.ycombinator.com/), where user-submitted stories (known as \"posts\") are voted and commented upon, similar to reddit. Hacker News is extremely popular in technology and startup circles, and posts that make it to the top of Hacker News' listings can get hundreds of thousands of visitors as a result.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We're specifically interested in posts whose titles begin with either `Ask HN` or `Show HN`. \n",
    "\n",
    "Users submit `Ask HN` posts to ask the Hacker News community a specific question. \n",
    "\n",
    "Below are a couple examples:\n",
    "\n",
    "\n",
    ">- Ask HN: What is that one deciding factor that makes a website successful?\n",
    "- Ask HN: How do you monitor server logs\n",
    "- Ask HN: What are the potential pitfalls of stock options?\n",
    "\n",
    "\n",
    "Likewise, users submit `Show HN` posts to show the Hacker News community a project, product, or just generally something interesting. \n",
    "\n",
    "Below are a couple of examples:\n",
    "\n",
    ">- Show HN: A simple library for complicated animations\n",
    "- Show HN: Hangman-game in command line. Knowing words is fun\n",
    "- Show HN: World's first kids art gallery\n",
    "\n",
    "\n",
    "We'll compare these two types of posts to determine the following:\n",
    "\n",
    "\n",
    "- **Do `Ask HN` or `Show HN` receive more comments on average?**\n",
    "- **Do posts created at a certain time receive more comments on average?**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## About the Dataset Used in This Project\n",
    "---\n",
    "The [data set](https://www.kaggle.com/hacker-news/hacker-news-posts) can be downloaded from this [link](https://www.kaggle.com/hacker-news/hacker-news-posts).\n",
    "\n",
    "This data set is Hacker News posts for 12 months (up to September 26 2016).\n",
    "\n",
    "It includes the following columns:\n",
    "\n",
    "- **title**: title of the post (self explanatory)\n",
    "\n",
    "- **url**: the url of the item being linked to\n",
    "\n",
    "- **num_points**: the number of upvotes the post received\n",
    "\n",
    "- **num_comments**: the number of comments the post received\n",
    "\n",
    "- **author**: the name of the account that made the post\n",
    "\n",
    "- **created_at**: the date and time the post was made (the time zone is Eastern Time in the US)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "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>id</th>\n",
       "      <th>title</th>\n",
       "      <th>url</th>\n",
       "      <th>num_points</th>\n",
       "      <th>num_comments</th>\n",
       "      <th>author</th>\n",
       "      <th>created_at</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>265909</td>\n",
       "      <td>10376312</td>\n",
       "      <td>Helpcrowd, questions from potential customers ...</td>\n",
       "      <td>http://www.helpcrowd.co/</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>cambridge</td>\n",
       "      <td>2015-10-12 18:47:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>157553</td>\n",
       "      <td>11214655</td>\n",
       "      <td>Franz  a free messaging app</td>\n",
       "      <td>http://meetfranz.com/</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>anu_gupta</td>\n",
       "      <td>2016-03-03 02:29:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>226375</td>\n",
       "      <td>10666951</td>\n",
       "      <td>Crazy Like a Box: Going Public Can Give Startu...</td>\n",
       "      <td>http://www.nytimes.com/2015/12/03/technology/c...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>hvo</td>\n",
       "      <td>2015-12-03 00:33:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>96869</td>\n",
       "      <td>11724506</td>\n",
       "      <td>Israel's LiveU to beam Summer Olympics events ...</td>\n",
       "      <td>http://www.timesofisrael.com/israels-liveu-to-...</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>jonbaer</td>\n",
       "      <td>2016-05-18 18:35:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>77205</td>\n",
       "      <td>11899729</td>\n",
       "      <td>Machine Intelligence Will Let Us All Work Like...</td>\n",
       "      <td>https://hbr.org/2016/06/machine-intelligence-w...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>scg</td>\n",
       "      <td>2016-06-14 04:14:00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              id                                              title  \\\n",
       "265909  10376312  Helpcrowd, questions from potential customers ...   \n",
       "157553  11214655                        Franz  a free messaging app   \n",
       "226375  10666951  Crazy Like a Box: Going Public Can Give Startu...   \n",
       "96869   11724506  Israel's LiveU to beam Summer Olympics events ...   \n",
       "77205   11899729  Machine Intelligence Will Let Us All Work Like...   \n",
       "\n",
       "                                                      url  num_points  \\\n",
       "265909                           http://www.helpcrowd.co/           1   \n",
       "157553                              http://meetfranz.com/           2   \n",
       "226375  http://www.nytimes.com/2015/12/03/technology/c...           1   \n",
       "96869   http://www.timesofisrael.com/israels-liveu-to-...           2   \n",
       "77205   https://hbr.org/2016/06/machine-intelligence-w...           1   \n",
       "\n",
       "        num_comments     author          created_at  \n",
       "265909             1  cambridge 2015-10-12 18:47:00  \n",
       "157553             1  anu_gupta 2016-03-03 02:29:00  \n",
       "226375             0        hvo 2015-12-03 00:33:00  \n",
       "96869              0    jonbaer 2016-05-18 18:35:00  \n",
       "77205              0        scg 2016-06-14 04:14:00  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "hn = pd.read_csv('HN_posts_year_to_Sep_26_2016.csv', parse_dates=['created_at'])\n",
    "\n",
    "### Preview 5 sample records from the dataset.\n",
    "\n",
    "hn.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 293119 entries, 0 to 293118\n",
      "Data columns (total 7 columns):\n",
      "id              293119 non-null int64\n",
      "title           293119 non-null object\n",
      "url             279256 non-null object\n",
      "num_points      293119 non-null int64\n",
      "num_comments    293119 non-null int64\n",
      "author          293119 non-null object\n",
      "created_at      293119 non-null datetime64[ns]\n",
      "dtypes: datetime64[ns](1), int64(3), object(3)\n",
      "memory usage: 15.7+ MB\n"
     ]
    }
   ],
   "source": [
    "### Check the dataset info... it looks good as there's no missing value in any column.\n",
    "\n",
    "hn.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(80401, 7)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Since we're only interested in records with comments, remove rows where number of comments is 0.\n",
    "\n",
    "hn = hn[hn[\"num_comments\"] > 0].copy()\n",
    "hn.shape\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (1) `Ask HN` and `Show HN` Posts: Which One Receive More Comments on Average?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "### This is a user-defined function to classify the type of post (i.e. 'Ask HN', 'Show HN', 'Other').\n",
    "\n",
    "def get_type(str):\n",
    "    \n",
    "    if str.lower().startswith(\"ask hn\"):\n",
    "        return \"Ask HN\"\n",
    "    elif str.lower().startswith(\"show hn\"):\n",
    "        return \"Show HN\"\n",
    "    else:\n",
    "        return \"Other\"\n",
    "   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>id</th>\n",
       "      <th>title</th>\n",
       "      <th>url</th>\n",
       "      <th>num_points</th>\n",
       "      <th>num_comments</th>\n",
       "      <th>author</th>\n",
       "      <th>created_at</th>\n",
       "      <th>type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>142193</td>\n",
       "      <td>11341553</td>\n",
       "      <td>Show HN: Nestor  Hosted Hubot for Slack</td>\n",
       "      <td>https://www.asknestor.me</td>\n",
       "      <td>49</td>\n",
       "      <td>11</td>\n",
       "      <td>iamclovin</td>\n",
       "      <td>2016-03-23 01:56:00</td>\n",
       "      <td>Show HN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>52577</td>\n",
       "      <td>12111508</td>\n",
       "      <td>How a Danish Town Helped Young Muslims Turn Aw...</td>\n",
       "      <td>http://www.npr.org/sections/health-shots/2016/...</td>\n",
       "      <td>88</td>\n",
       "      <td>87</td>\n",
       "      <td>tim_sw</td>\n",
       "      <td>2016-07-17 19:54:00</td>\n",
       "      <td>Other</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>96609</td>\n",
       "      <td>11727106</td>\n",
       "      <td>Theranos voids two years of Edison blood-test ...</td>\n",
       "      <td>http://www.wsj.com/articles/theranos-voids-two...</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>whbk</td>\n",
       "      <td>2016-05-19 00:26:00</td>\n",
       "      <td>Other</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>117869</td>\n",
       "      <td>11545623</td>\n",
       "      <td>Ask HN: Rough estimate on how long it can take...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>throwawaywebdev</td>\n",
       "      <td>2016-04-21 21:58:00</td>\n",
       "      <td>Ask HN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>45108</td>\n",
       "      <td>12176410</td>\n",
       "      <td>Show HN: VR demo of living the same minute ove...</td>\n",
       "      <td>https://www.youtube.com/watch?v=_eoMTN509P8</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>iamwil</td>\n",
       "      <td>2016-07-27 20:48:00</td>\n",
       "      <td>Show HN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              id                                              title  \\\n",
       "142193  11341553            Show HN: Nestor  Hosted Hubot for Slack   \n",
       "52577   12111508  How a Danish Town Helped Young Muslims Turn Aw...   \n",
       "96609   11727106  Theranos voids two years of Edison blood-test ...   \n",
       "117869  11545623  Ask HN: Rough estimate on how long it can take...   \n",
       "45108   12176410  Show HN: VR demo of living the same minute ove...   \n",
       "\n",
       "                                                      url  num_points  \\\n",
       "142193                           https://www.asknestor.me          49   \n",
       "52577   http://www.npr.org/sections/health-shots/2016/...          88   \n",
       "96609   http://www.wsj.com/articles/theranos-voids-two...          10   \n",
       "117869                                                NaN           6   \n",
       "45108         https://www.youtube.com/watch?v=_eoMTN509P8           1   \n",
       "\n",
       "        num_comments           author          created_at     type  \n",
       "142193            11        iamclovin 2016-03-23 01:56:00  Show HN  \n",
       "52577             87           tim_sw 2016-07-17 19:54:00    Other  \n",
       "96609              3             whbk 2016-05-19 00:26:00    Other  \n",
       "117869             5  throwawaywebdev 2016-04-21 21:58:00   Ask HN  \n",
       "45108              1           iamwil 2016-07-27 20:48:00  Show HN  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Create a new column to indicate the type of post (i.e. 'Ask HN', 'Show HN', 'Other')\n",
    "\n",
    "hn[\"type\"] = hn[\"title\"].apply(get_type)\n",
    "hn.sample(5)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Since we're only interested in 'Ask HN' and 'Show HN' posts, let's remove the others.\n",
    "\n",
    "hn_ask_show = hn[hn[\"type\"] != \"Other\"].copy()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ask HN     6911\n",
       "Show HN    5059\n",
       "Name: type, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### The number of 'Ask HN' and 'Show HN' posts.\n",
    "\n",
    "hn_ask_show[\"type\"].value_counts()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Generate bar plot for the number of 'Ask HN' and 'Show HN' posts.\n",
    "\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('fivethirtyeight')\n",
    "\n",
    "hn_ask_show[\"type\"].value_counts().plot.bar(title=\"No. of Posts\", rot=0)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "type\n",
       "Ask HN     13.744176\n",
       "Show HN     9.810832\n",
       "Name: num_comments, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Getting the average number of comments by the type of post.\n",
    "\n",
    "hn_avg_comments = hn_ask_show.groupby(\"type\")[\"num_comments\"].mean().sort_values(ascending=False)\n",
    "hn_avg_comments\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Generate bar plot for the average number of comments by the type of post.\n",
    "\n",
    "plt.style.use('fivethirtyeight')\n",
    "hn_avg_comments.plot.bar(title=\"Avg No. of Comments\", legend=False, rot=0)\n",
    "plt.ylabel(\"\")\n",
    "plt.xlabel(\"\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Observations : On average, `Ask HN` posts receive more comments than `Show HN` posts.**\n",
    "\n",
    "- about 14 comments for `Ask HN` posts\n",
    "- about 10 comments for `Show HN` posts\n",
    "\n",
    "Next Step: Since `Ask HN` posts are more likely to receive comments, we'll focus our remaining analysis just on these posts.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (2) Getting The Average Number of Comments for `Ask HN` Posts by Hour\n",
    "\n",
    "- Calculate the amount of `Ask HN` posts created in each hour of the day, along with the number of comments received.\n",
    "- Calculate the average number of comments `Ask HN` posts receive by hour created."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "hn_ask = hn_ask_show[hn_ask_show[\"type\"] == \"Ask HN\"].copy()\n",
    "\n",
    "### Creating a new column to indicate the hour the 'Ask HN' post is created.\n",
    "\n",
    "hn_ask[\"created_hour\"] = hn_ask[\"created_at\"].dt.hour\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15    467\n",
       "18    452\n",
       "19    420\n",
       "16    415\n",
       "21    407\n",
       "17    404\n",
       "20    392\n",
       "14    378\n",
       "13    326\n",
       "22    287\n",
       "23    276\n",
       "12    274\n",
       "11    251\n",
       "0     231\n",
       "2     227\n",
       "1     223\n",
       "10    219\n",
       "3     212\n",
       "8     190\n",
       "4     186\n",
       "6     176\n",
       "9     176\n",
       "5     165\n",
       "7     157\n",
       "Name: created_hour, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Number of 'Ask HN' posts created in each hour of the day.\n",
    "\n",
    "hn_ask[\"created_hour\"].value_counts()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks like most `Ask HN` posts are created between 3pm to 9pm (on US Eastern Time)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "created_hour\n",
       "0      9.857143\n",
       "1      9.367713\n",
       "2     13.198238\n",
       "3     10.160377\n",
       "4     12.688172\n",
       "5     11.139394\n",
       "6      9.017045\n",
       "7     10.095541\n",
       "8     12.431579\n",
       "9      8.392045\n",
       "10    13.757991\n",
       "11    11.143426\n",
       "12    15.452555\n",
       "13    22.223926\n",
       "14    13.153439\n",
       "15    39.668094\n",
       "16    10.761446\n",
       "17    13.730198\n",
       "18    10.789823\n",
       "19     9.414286\n",
       "20    11.382653\n",
       "21    11.056511\n",
       "22    11.749129\n",
       "23     8.322464\n",
       "Name: num_comments, dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### Getting the average number of comments for 'Ask HN' posts by hour of the day.\n",
    "\n",
    "ask_avg_comments = hn_ask.groupby(\"created_hour\")[\"num_comments\"].mean()\n",
    "ask_avg_comments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Generate line plot for the average number of comments for Ask HN posts created during the specific hours.\n",
    "\n",
    "#plt.style.use('fivethirtyeight')\n",
    "ask_avg_comments.plot(title=\"Avg No. of Comments Received\\n for 'Ask HN' Posts\", legend=False, rot=0)\n",
    "plt.ylabel(\"\")\n",
    "plt.xlabel(\"Creation Time (Hour of the day)\", fontsize = 12)\n",
    "plt.xticks(list(range(0,24,2)))\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "created_hour\n",
       "15    39.668094\n",
       "13    22.223926\n",
       "12    15.452555\n",
       "10    13.757991\n",
       "17    13.730198\n",
       "Name: num_comments, dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ask_avg_comments_top5 = hn_ask.groupby(\"created_hour\")[\"num_comments\"].mean().sort_values(ascending=False).head(5)\n",
    "ask_avg_comments_top5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**From the result and graph above, we can see that the top 5 hours of the day (on US Eastern Time) where `Ask HN` posts was created and receive the most comments are:**\n",
    "\n",
    "- 15:00 (3pm) -> 39.67 average comments per post\n",
    "- 13:00 (1pm) -> 22.22 average comments per post\n",
    "- 12:00 (12pm) -> 15.45 average comments per post\n",
    "- 10:00 (10am) -> 13.76 average comments per post\n",
    "- 17:00 (5pm) -> 13.73 average comments per post\n",
    "\n",
    "`Ask HN` posts created during these hours seems to receive the most comments."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Convert the times to the local time zone\n",
    "\n",
    "The records in this dataset are on US Eastern Time.\n",
    "\n",
    "- US Eastern Time Zone is GMT/UTC - 5h during Standard Time.\n",
    "- US Eastern Time Zone is GMT/UTC - 4h during Daylight Saving Time.\n",
    "- Singapore is GMT/UTC + 8h during Standard Time.\n",
    "\n",
    "US Eastern Time is 13 hours earlier than Singapore during Standard Time.\n",
    "\n",
    "Example: `Tuesday 2/4/2020 1:52 AM EST` = `Tuesday 2/4/2020 2:52 PM UTC+08:00`\n",
    "\n",
    "Therefore, to convert the time to SGT time, we need to add 13 hours."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "created_hour_SGT\n",
       "4     39.668094\n",
       "2     22.223926\n",
       "1     15.452555\n",
       "23    13.757991\n",
       "6     13.730198\n",
       "Name: num_comments, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import datetime as dt\n",
    "\n",
    "### Creating a new column to indicate the hour the 'Ask HN' post is created at SGT local time.\n",
    "\n",
    "hn_ask[\"created_hour_SGT\"] = hn_ask[\"created_at\"] + dt.timedelta(hours=13)\n",
    "hn_ask[\"created_hour_SGT\"] = hn_ask[\"created_hour_SGT\"].dt.hour\n",
    "\n",
    "### Getting the average number of comments for 'Ask HN' posts by hour, on SGT local time.\n",
    "\n",
    "hn_ask.groupby(\"created_hour_SGT\")[\"num_comments\"].mean().sort_values(ascending=False).head(5)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**After conversion, we can see that the top 5 hours of the day (on SGT Local Time) where `Ask HN` posts was created and receive the most comments are:**\n",
    "\n",
    "- 04:00 (4am) -> 39.67 average comments per post\n",
    "- 02:00 (2am) -> 22.22 average comments per post\n",
    "- 01:00 (1am) -> 15.45 average comments per post\n",
    "- 23:00 (11pm) -> 13.76 average comments per post\n",
    "- 06:00 (6am) -> 13.73 average comments per post\n",
    "\n",
    "`Ask HN` posts created during these hours seems to receive the most comments."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (3) Conclusion\n",
    "\n",
    "Which hours should you create a post during the day to have a higher chance of receiving comments?\n",
    "\n",
    "Well, a post created during the following time would have a higher chance of receiving comments:\n",
    "\n",
    "- Between 10am to 5pm on US Eastern Time (the peak is at 3pm).\n",
    "- Between 11pm to 6am on SGT local time (the peak is at 4am)."
   ]
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}