act_report.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Analysis Report"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Reporting Requirements:\n",
    "\n",
    "- Create a **250-word-minimum written report** called *act_report.pdf or act_report.html* that communicates the insights and displays the visualization(s) produced from your wrangled data. This is to be framed as an external document, like a blog post or magazine article, for example."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Summary of questions\n",
    "\n",
    "*During the data wrangling process, I came up with 4 questions that I thought would be worth-exploring:*\n",
    "\n",
    ">1. What is the relationship between favourite_count and retweet_count on Twitter?\n",
    ">2. What are the sources of the tweets? \n",
    ">3. From the result of the first algorithm to predict the breed of the dog via images, how many images can be predicted with at least 95% confidence and what are they top 5 breeds from the prediction? \n",
    ">4. Among the 3 prediction algorithms, which one is more reliable?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Question 1: What is the relationship between favourite_count and retweet_count on Twitter?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The correlation coefficient between retweet and favourite count is 0.861987708294\n"
     ]
    }
   ],
   "source": [
    "# We want to understand the relationship between retweet and favourite count"
   ]
  },
  {
   "attachments": {
    "test1.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![test1.png](attachment:test1.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">#### Insight 1(a)\n",
    "- The scatter plot above shows that favourite counts and retweet counts are positively correlated. \n",
    "- The correlation coefficient between retweet and favourite counts is 0.86, showing a strong positively correlated relationship.\n",
    "- Some retweets have no favorite_count. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>     <td>favorite_count</td>  <th>  R-squared:         </th> <td>   0.743</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.743</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   5948.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Wed, 04 Mar 2020</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:36:28</td>     <th>  Log-Likelihood:    </th> <td> -20892.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>  2059</td>      <th>  AIC:               </th> <td>4.179e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>  2057</td>      <th>  BIC:               </th> <td>4.180e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "        <td></td>           <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>intercept</th>     <td> 2054.6223</td> <td>  156.851</td> <td>   13.099</td> <td> 0.000</td> <td> 1747.019</td> <td> 2362.225</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>retweet_count</th> <td>    2.2989</td> <td>    0.030</td> <td>   77.121</td> <td> 0.000</td> <td>    2.240</td> <td>    2.357</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>1212.289</td> <th>  Durbin-Watson:     </th> <td>   1.182</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th>  <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td>88439.298</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>           <td>-1.951</td>  <th>  Prob(JB):          </th> <td>    0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>       <td>34.869</td>  <th>  Cond. No.          </th> <td>6.06e+03</td> \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:         favorite_count   R-squared:                       0.743\n",
       "Model:                            OLS   Adj. R-squared:                  0.743\n",
       "Method:                 Least Squares   F-statistic:                     5948.\n",
       "Date:                Wed, 04 Mar 2020   Prob (F-statistic):               0.00\n",
       "Time:                        11:36:28   Log-Likelihood:                -20892.\n",
       "No. Observations:                2059   AIC:                         4.179e+04\n",
       "Df Residuals:                    2057   BIC:                         4.180e+04\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "=================================================================================\n",
       "                    coef    std err          t      P>|t|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------\n",
       "intercept      2054.6223    156.851     13.099      0.000    1747.019    2362.225\n",
       "retweet_count     2.2989      0.030     77.121      0.000       2.240       2.357\n",
       "==============================================================================\n",
       "Omnibus:                     1212.289   Durbin-Watson:                   1.182\n",
       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):            88439.298\n",
       "Skew:                          -1.951   Prob(JB):                         0.00\n",
       "Kurtosis:                      34.869   Cond. No.                     6.06e+03\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 6.06e+03. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The statistics below tell us more whether it is useful to use retweet as an indepedent \n",
    "# variable affecting the number of favourite counts on Twitter"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">#### Insight (1b)\n",
    "- In the regression model above, retweet_count is the independent variable and the favourite_count is the dependent variable. \n",
    "- The regression model above is able to explain 74.3% of the population.\n",
    "- The P-value of the retweet_count is 0.00, which means that we can reject the null hypothesis of retweet_count having no effect on favourite_count. \n",
    "- On average, every one unit of retweet_count leads to an increase of favourite_count by 2.298 unit. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Question 2: What are the sources of the tweets? "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Sources and the counts:**\n",
    "- Twitter for iPhone : 2019\n",
    "<br>\n",
    "- Twitter Web Client : 30\n",
    "<br>\n",
    "- TweetDeck : 10\n"
   ]
  },
  {
   "attachments": {
    "test2.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![test2.png](attachment:test2.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**Insight 2**\n",
    "- The tweets are predominantly from iPhones.\n",
    "- There are barely any users from the Twitter website or TweetDeck. \n",
    "- However, it seems like the data is lacking information about users who are not using iPhone. There could be a chance that all mobile users have been grouped up as iPhone but this is worth investigating. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Question 3: From the result of the first algorithm to predict the breed of the dog via images, how many images can be predicted with at least 95% confidence and what are they top 5 breeds from the prediction? "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**Insight 3**\n",
    "- From the 1st algorithm predicting the breed of dogs, only 188 images could be predicted with at least 95% confidence. \n",
    "- When dogs are holding other objects, it seems like all the predictions won't be able to differentiate between objects and dogs from the image as each of the 3 algorithms has predicted at least 500 non-dog objects.\n",
    "- Only about 9% of images can be predicted with at least 95% confidence.\n",
    "- The top 5 breeds from the prediction are pug (21), samoyed (19),pembroke (19), golden_retriever (18) and pomeranian (13)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Question 4: Among the 3 prediction algorithms, which one is more reliable?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">**Insight 4**\n",
    "- Among the 3 algorithms, it seems that the first algorithm is more reliable in predicting the breed of dogs via its images as it is able to predict 9% of images with at least 95% confidence, while the other two algorithms do have any prediction with more than 95% confidence."
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Attachments",
  "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}