{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<font style='font-size:3em'>**📝 Week 5 Summative Assessment** </font>\n",
    "\n",
    "## DS105A – Data for Data Science\n",
    "\n",
    "**PURPOSE**: The purpose of this Jupyter Notebook is to document my answers to the DS105 W5 Summative assessment, show the steps I took and explain the rationale behind these decisions. It will also include some additional, relevant insights into the data itself.\n",
    "\n",
    "**CLICK THE IMAGE BELOW TO VIEW THE WEBSITE THAT WAS SCRAPED FOR THIS PROJECT:**\n",
    "\n",
    "<a href=\"https://socialdatascience.network/index.html#intro\">\n",
    "    <img src=\"https://hertieschool-f4e6.kxcdn.com/fileadmin/_processed_/7/b/csm_2022_CIVICA_Website_white_180x90_f9be83fbe7.png\" alt=\"Civica logo\" width=\"25%\">\n",
    "</a>\n",
    "\n",
    "**LAST REVISION:** 30th October 2023"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:70%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "Notice how well organise this notebook is! It even has pictures. It's easy to read and follow. It's also easy to see what the author has done and why. This is a great example of how to present your work.\n",
    "\n",
    "Another way to have a sense for what is in a notebook is by clicking on the **Outline** button above, while the notebook is open in VS Code. This will show you all the headings in the notebook. You can then click on a heading to jump to that section of the notebook.\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "## ⚙️ Setting Up\n",
    "\n",
    "### Packages Downloaded and Updates Made\n",
    "\n",
    "- The following commands were run from the terminal within VS Code:\n",
    "\n",
    "    ```console\n",
    "    pip install pandas\n",
    "    pip install requests\n",
    "    sudo pip3 install requests\n",
    "    pip3 install --upgrade pip\n",
    "    pip install scrapy\n",
    "    sudo pip3 install scrapy\n",
    "    pip install ordered-set\n",
    "    pip install selenium\n",
    "    pip install nltk\n",
    "    pip install certifi\n",
    "    pip install matplotlib\n",
    "    pip install plotly\n",
    "    pip install --upgrade nbformat\n",
    "    pip install gensim\n",
    "    pip install --upgrade gensim\n",
    "    pip install inflect\n",
    "    ```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:70%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "This is a fantastic great practice! I know exactly what I need to type on the Terminal to get the same packages they used in the submission\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importing Installed Packages\n",
    "\n",
    "- The below chunk of code was run to import the installed packages\n",
    "- Python 3.11.5 is being used"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests            \n",
    "import pandas as pd        \n",
    "from pprint import pprint    \n",
    "from scrapy import Selector  \n",
    "from ordered_set import OrderedSet  \n",
    "from collections import Counter\n",
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "import certifi\n",
    "import os\n",
    "from datetime import datetime"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:70%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "- This person discovered these very useful Python packages (`ordered_set` and `collections`). It is not clear if those were discovered with the help of generative AI (they should have ellaborated a bit more on this at the end), but they used it well.\n",
    "\n",
    "-  This person used other Python packages useful for text mining analysis (!) This made us go WOW! We weren't expecting any deep analysis this early in the course.\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Folders Created\n",
    "- `Data` folder is created using a relative path (so that my username/name isn't given away by the path)\n",
    "- This contains `Schedule.CSV` (from Part 1 of the Summative) and `Agenda.CSV` (from Part 2 of the Summative), these files are both saved directly to the `Data` Folder\n",
    "- An `if` statement is used so that multiple `Data` folders are not generated, and whenever this Jupyter Notebook is ran the new `agenda.csv` and `schedule.csv` will replace the old version within the `Data` folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Folder 'Data' already exists in the current directory.\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "# Get the current working directory (where your Python script is located)\n",
    "current_directory = os.getcwd()\n",
    "\n",
    "# Specify the name of your new folder\n",
    "folder_name = 'Data'\n",
    "\n",
    "# Construct the relative path to the new folder\n",
    "data_folder_path = os.path.join(current_directory, folder_name)\n",
    "\n",
    "# Check if the folder doesn't exist already, then create it\n",
    "if not os.path.exists(data_folder_path):\n",
    "    os.makedirs(data_folder_path)\n",
    "    print(f\"Folder '{folder_name}' created successfully in the current directory.\")\n",
    "else:\n",
    "    print(f\"Folder '{folder_name}' already exists in the current directory.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:70%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "This is great! They found a way to make this replicable and reproducible. The [`os` library](https://docs.python.org/3/library/os.html), from the standard Python library, is used to create a folder and save the files there, emulating what one would do in the Terminal. This is a very good practice.\n",
    "\n",
    "If you aren't familiar with the `os` library, you could have achieved similar greatness by adding a markdown cell with instructions: \n",
    "\n",
    "> Go to the Terminal and type the following before running the rest of this notebook\n",
    ">\n",
    "> ```console\n",
    "> mkdir Data\n",
    "> ```\n",
    ">\n",
    "\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "## 🔐 Requesting a Web Page\n",
    "\n",
    "### Requesting the Website and using a Selector\n",
    "\n",
    "- We store the URL of the target website, send a GET request to the specified URL and store the server's response\n",
    "- We then create a Scrapy Selector object which allows us to apply CSS selectors to extract specific data from the HTML code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# This is the address of the website we want to scrape\n",
    "my_url = 'https://socialdatascience.network/index.html#schedule'\n",
    "\n",
    "# We set a GET request to the website\n",
    "response = requests.get(my_url)\n",
    "\n",
    "# Parse the HTML code using Scrapy Selector\n",
    "sel = Selector(text=response.text)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "## 🔍 Part 1: Scraping for titles, speakers, dates and links to CIVICA Seminar Schedule\n",
    "\n",
    "### Scrape the **names of all the events** and save them to a list\n",
    "\n",
    "- All event titles are represented within a `<h6 class=\"card-title\">` tag (found by inspecting the page)\n",
    "- Therefore: `h6.card-title` is a good CSS Selector to use"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saving all the event titles as a list\n",
    "titles = sel.css('h6.card-title ::text').getall()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scrape the **links to all the events** and save them to a list\n",
    "\n",
    "- From inspecting the page we can see that all event links are under the attribute `href` and are represented within `<div class=\"card-body\">`\n",
    "- Therefore: `div.card-body a ::attr(href)` is a good selector to use\n",
    "- The generated list has duplicates, so to remove these I used an ordered set (a normal set would have changed the order of the links leading to mismatched data in the final table). Although I considered using a for loop to remove duplicates, I found that the ordered set was much simpler\n",
    "- I also added the prefix of \"https://socialdatascience.network/\" to all the links to make them full links instead of partial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saving all the links to events as a list\n",
    "from ordered_set import OrderedSet\n",
    "partial_links = OrderedSet(sel.css('div.card-body a ::attr(href)').getall())\n",
    "\n",
    "# Adding a prefix to all the links in the list\n",
    "address = \"https://socialdatascience.network/\"                             \n",
    "links = []\n",
    "for partial_link in partial_links:\n",
    "    full_link = address + partial_link\n",
    "    links.append(full_link)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:80%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "Their solution is elegant in its use of the `ordered_set` package to remove duplicated entries. If you were dealing with thousand, or millions, of entries, this would be a very efficient way to remove duplicates.\n",
    "\n",
    "But even without knowledge of this package, you could have achieved the same result by using a regular, slightly less efficient, Python `set` data structure. A concise alternative to the chunk above would be to use list comprehension:\n",
    "\n",
    "```python\n",
    "partial_links = set(sel.css('div.card-body a ::attr(href)').getall())\n",
    "links = [address+partial_link for partial_link in partial_links]\n",
    "```\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scrape the **names of all the speakers** and **dates of all the events** and save them to seperate lists\n",
    "\n",
    "- All speakers names are represented within a `<p class=\"card-text\">` tag which would be our ideal CSS selector\n",
    "- However, both the speaker name and date are contained within this, separated by a `<br>`\n",
    "- We can separate the speaker name from the date of the event using a for loop, and ammend different lists to keep the names and dates in separate lists"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "speakers = []\n",
    "dates = []\n",
    "\n",
    "for i in sel.css('div.card-body p ::text').getall():\n",
    "    i = i.lstrip()\n",
    "    if i.startswith('Speaker:'):\n",
    "        speaker_name = i.replace('Speaker: ', '')\n",
    "        speakers.append(speaker_name)\n",
    "    elif i.startswith('Date:'):\n",
    "        date = i.replace('Date: ', '')\n",
    "        dates.append(date)\n",
    "    else:\n",
    "        speakers.append('')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Convert the lists to a **pandas data frame** and save it to a **CSV file**\n",
    "\n",
    "- Note: In order to use the `pd.DataFrame()` function to create a data frame all the lists/arrays must be the same length. This was checked using `if` statement\n",
    "- The final CSV file (schedule.csv) is saved directly to the data folder creating in the Setting Up stage\n",
    "- The final CSV file (schedule.csv) can be viewed as a table\n",
    "- If an event has no speakers, then the relevant cell of the table will be left empty"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All lists are same length\n"
     ]
    },
    {
     "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>title</th>\n",
       "      <th>speaker</th>\n",
       "      <th>date</th>\n",
       "      <th>link</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Promoting the systematic use of real-world dat...</td>\n",
       "      <td>Dr. Divya Srivastava, LSE</td>\n",
       "      <td>Wednesday, 22 November 2023</td>\n",
       "      <td>https://socialdatascience.network/fall2023/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Data science for the Sustainable Development G...</td>\n",
       "      <td>Prof. Elisa Omodei, CEU</td>\n",
       "      <td>Wednesday, 18 October 2023</td>\n",
       "      <td>https://socialdatascience.network/fall2023/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CentralBankRoBERTa: A Fine-Tuned Large Languag...</td>\n",
       "      <td>Moritz Pfeifer &amp; Vincent Philipp Marohl</td>\n",
       "      <td>Wednesday, 27 September 2023</td>\n",
       "      <td>https://socialdatascience.network/fall2023/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>The Evolution of the Climate Discourse on Twit...</td>\n",
       "      <td>Dr. Max Falkenberg</td>\n",
       "      <td>Wednesday, 13 September 2023</td>\n",
       "      <td>https://socialdatascience.network/fall2023/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>The Handbook of Computational Social Science f...</td>\n",
       "      <td>Dr. Eleonora Bertoni</td>\n",
       "      <td>Wednesday, 31 May 2023</td>\n",
       "      <td>https://socialdatascience.network/spring2023/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Artificial Intelligence, Algorithmic Recommend...</td>\n",
       "      <td>Prof. Giacomo Calzolari</td>\n",
       "      <td>Wednesday, 03 May 2023</td>\n",
       "      <td>https://socialdatascience.network/spring2023/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Exploring A New Model of Industry/Academic Col...</td>\n",
       "      <td>Prof. Pablo Barberá</td>\n",
       "      <td>Wednesday, 19 April 2023</td>\n",
       "      <td>https://socialdatascience.network/spring2023/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Using Multimodal Neural Networks to Better Und...</td>\n",
       "      <td>Prof. Bryce Jensen Dietrich</td>\n",
       "      <td>Wednesday, 22 March 2023</td>\n",
       "      <td>https://socialdatascience.network/spring2023/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Models, mathematics, and data science: how to ...</td>\n",
       "      <td>Dr. Erica Thompson</td>\n",
       "      <td>Wednesday, 08 March 2023</td>\n",
       "      <td>https://socialdatascience.network/spring2023/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>CIVICA Conference on European Polarisation</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Wednesday, 15 February 2023</td>\n",
       "      <td>https://socialdatascience.network/polarisation...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>New Faces of Bias in Online Platforms</td>\n",
       "      <td>Prof. Aniko Hannak</td>\n",
       "      <td>Wednesday, 08 February 2023</td>\n",
       "      <td>https://socialdatascience.network/spring2023/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Introducing the Online Harms Observatory: AI p...</td>\n",
       "      <td>Pica Johnsson</td>\n",
       "      <td>Wednesday, 11 January 2023</td>\n",
       "      <td>https://socialdatascience.network/spring2023/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Using Open Source Data Streams and Surveys to ...</td>\n",
       "      <td>Prof. Lisa Singh</td>\n",
       "      <td>Wednesday, 02 November 2022</td>\n",
       "      <td>https://socialdatascience.network/fall2022/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Does Epistemic Vice Explain Corporate Misconduct?</td>\n",
       "      <td>Dr. Marco Meyer</td>\n",
       "      <td>Wednesday, 19 October 2022</td>\n",
       "      <td>https://socialdatascience.network/fall2022/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Becoming a data scientist: what it means to pu...</td>\n",
       "      <td>Prof. Anne Beaulieu</td>\n",
       "      <td>Wednesday, 14 September 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>The Making of a French Migration Crisis</td>\n",
       "      <td>Dr. Michelle Reddy &amp; Dr. Hélène Thiollet</td>\n",
       "      <td>Wednesday, 15 June 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>A New Approach to Visualizing Spatial Exposure...</td>\n",
       "      <td>Prof. Stephanie Lackner</td>\n",
       "      <td>Wednesday, 01 June 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Modeling Sustainable Development from the Bott...</td>\n",
       "      <td>Dr. Omar A. Guerrero</td>\n",
       "      <td>Wednesday, 04 May 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Internet Communities and the French Presidenti...</td>\n",
       "      <td>Prof. David Chavalarias</td>\n",
       "      <td>Wednesday, 20 April 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>The Science of Success: Quantifying Outcomes i...</td>\n",
       "      <td>Prof. Laszlo Barabasi</td>\n",
       "      <td>Wednesday, 09 March 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Embedding Regression: Models for Context-Speci...</td>\n",
       "      <td>Prof. Arthur Spirling</td>\n",
       "      <td>Wednesday, 23 February 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Information and Irregular Migration: Evidence ...</td>\n",
       "      <td>Dr. Alexandra Scacco</td>\n",
       "      <td>Wednesday, 09 February 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Adjusting for Confounding with Text Matching</td>\n",
       "      <td>Prof. Margaret Roberts</td>\n",
       "      <td>Wednesday, 26 January 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>What is Data Feminist?</td>\n",
       "      <td>Prof. Lauren Klein</td>\n",
       "      <td>Wednesday, 12 January 2022</td>\n",
       "      <td>https://socialdatascience.network/spring2022/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>The Principles of Collective Learning</td>\n",
       "      <td>Prof. Cesar A. Hidalgo</td>\n",
       "      <td>Wednesday, 1 December 2021</td>\n",
       "      <td>https://socialdatascience.network/fall2021/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>More Than Words: How Political Rhetoric Shapes...</td>\n",
       "      <td>Prof. Christopher Lucas</td>\n",
       "      <td>Wednesday, 3 November 2021</td>\n",
       "      <td>https://socialdatascience.network/fall2021/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Serendipity or Confinement? Deconstructing Alg...</td>\n",
       "      <td>Prof. Camille Roth</td>\n",
       "      <td>Wednesday, 20 October 2021</td>\n",
       "      <td>https://socialdatascience.network/fall2021/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Framing a Protest: Determinants and Effects of...</td>\n",
       "      <td>Prof. Michelle Torrest</td>\n",
       "      <td>Wednesday, 06 October 2021</td>\n",
       "      <td>https://socialdatascience.network/fall2021/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>Understanding Beautiful Places with AI</td>\n",
       "      <td>Prof. Suzy Moat</td>\n",
       "      <td>Wednesday, 22 September, 2021</td>\n",
       "      <td>https://socialdatascience.network/sess9.html</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Incentives and Covid-19 Vaccination Uptake</td>\n",
       "      <td>Prof. Macartan Humphreys</td>\n",
       "      <td>Wednesday, 08 September 2021</td>\n",
       "      <td>https://socialdatascience.network/fall2021/ses...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>The Art of Quantitative Editing</td>\n",
       "      <td>Dr. Laura Bronner</td>\n",
       "      <td>Wednesday, 02 June 2021</td>\n",
       "      <td>https://socialdatascience.network/sess7.html</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>Breaking the Social Media Prism</td>\n",
       "      <td>Chris Bail</td>\n",
       "      <td>Wednesday, 19 May, 2021</td>\n",
       "      <td>https://socialdatascience.network/sess6.html</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>Using Public Video Cameras to Detect Racial Di...</td>\n",
       "      <td>Dr. Melissa Sands</td>\n",
       "      <td>Wednesday, 05 May 2021</td>\n",
       "      <td>https://socialdatascience.network/sess5.html</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>How to Detect Fake News Before It Is Written</td>\n",
       "      <td>Dr. Preslav Nakov</td>\n",
       "      <td>Wednesday, 21 April 2021</td>\n",
       "      <td>https://socialdatascience.network/sess4.html</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>Negotiating with AI: Fairness in the Labor Market</td>\n",
       "      <td>Prof. Christo Wilson</td>\n",
       "      <td>Wednesday, 07 April, 2021</td>\n",
       "      <td>https://socialdatascience.network/sess3.html</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>Tracking Covid-19 with the Financial Times</td>\n",
       "      <td>John Burn-Murdoch</td>\n",
       "      <td>Wednesday, 24 March 2021</td>\n",
       "      <td>https://socialdatascience.network/sess2.html</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>Police Diversity to Prevent Violence: Does It ...</td>\n",
       "      <td>Roman Rivera</td>\n",
       "      <td>Wednesday, 17 March 2021</td>\n",
       "      <td>https://socialdatascience.network/sess1.html</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>Data Science in the Time of Covid and What Hap...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Wednesday, 24 February, 2021</td>\n",
       "      <td>https://socialdatascience.network/launch.html</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                title  \\\n",
       "0   Promoting the systematic use of real-world dat...   \n",
       "1   Data science for the Sustainable Development G...   \n",
       "2   CentralBankRoBERTa: A Fine-Tuned Large Languag...   \n",
       "3   The Evolution of the Climate Discourse on Twit...   \n",
       "4   The Handbook of Computational Social Science f...   \n",
       "5   Artificial Intelligence, Algorithmic Recommend...   \n",
       "6   Exploring A New Model of Industry/Academic Col...   \n",
       "7   Using Multimodal Neural Networks to Better Und...   \n",
       "8   Models, mathematics, and data science: how to ...   \n",
       "9          CIVICA Conference on European Polarisation   \n",
       "10              New Faces of Bias in Online Platforms   \n",
       "11  Introducing the Online Harms Observatory: AI p...   \n",
       "12  Using Open Source Data Streams and Surveys to ...   \n",
       "13  Does Epistemic Vice Explain Corporate Misconduct?   \n",
       "14  Becoming a data scientist: what it means to pu...   \n",
       "15            The Making of a French Migration Crisis   \n",
       "16  A New Approach to Visualizing Spatial Exposure...   \n",
       "17  Modeling Sustainable Development from the Bott...   \n",
       "18  Internet Communities and the French Presidenti...   \n",
       "19  The Science of Success: Quantifying Outcomes i...   \n",
       "20  Embedding Regression: Models for Context-Speci...   \n",
       "21  Information and Irregular Migration: Evidence ...   \n",
       "22       Adjusting for Confounding with Text Matching   \n",
       "23                             What is Data Feminist?   \n",
       "24              The Principles of Collective Learning   \n",
       "25  More Than Words: How Political Rhetoric Shapes...   \n",
       "26  Serendipity or Confinement? Deconstructing Alg...   \n",
       "27  Framing a Protest: Determinants and Effects of...   \n",
       "28             Understanding Beautiful Places with AI   \n",
       "29         Incentives and Covid-19 Vaccination Uptake   \n",
       "30                    The Art of Quantitative Editing   \n",
       "31                    Breaking the Social Media Prism   \n",
       "32  Using Public Video Cameras to Detect Racial Di...   \n",
       "33       How to Detect Fake News Before It Is Written   \n",
       "34  Negotiating with AI: Fairness in the Labor Market   \n",
       "35         Tracking Covid-19 with the Financial Times   \n",
       "36  Police Diversity to Prevent Violence: Does It ...   \n",
       "37  Data Science in the Time of Covid and What Hap...   \n",
       "\n",
       "                                      speaker                           date  \\\n",
       "0                  Dr. Divya Srivastava, LSE     Wednesday, 22 November 2023   \n",
       "1                    Prof. Elisa Omodei, CEU      Wednesday, 18 October 2023   \n",
       "2    Moritz Pfeifer & Vincent Philipp Marohl    Wednesday, 27 September 2023   \n",
       "3                         Dr. Max Falkenberg    Wednesday, 13 September 2023   \n",
       "4                       Dr. Eleonora Bertoni          Wednesday, 31 May 2023   \n",
       "5                    Prof. Giacomo Calzolari          Wednesday, 03 May 2023   \n",
       "6                        Prof. Pablo Barberá        Wednesday, 19 April 2023   \n",
       "7                Prof. Bryce Jensen Dietrich        Wednesday, 22 March 2023   \n",
       "8                         Dr. Erica Thompson        Wednesday, 08 March 2023   \n",
       "9                                         NaN    Wednesday, 15 February 2023   \n",
       "10                        Prof. Aniko Hannak     Wednesday, 08 February 2023   \n",
       "11                             Pica Johnsson      Wednesday, 11 January 2023   \n",
       "12                          Prof. Lisa Singh     Wednesday, 02 November 2022   \n",
       "13                           Dr. Marco Meyer      Wednesday, 19 October 2022   \n",
       "14                       Prof. Anne Beaulieu    Wednesday, 14 September 2022   \n",
       "15  Dr. Michelle Reddy & Dr. Hélène Thiollet         Wednesday, 15 June 2022   \n",
       "16                   Prof. Stephanie Lackner         Wednesday, 01 June 2022   \n",
       "17                      Dr. Omar A. Guerrero          Wednesday, 04 May 2022   \n",
       "18                   Prof. David Chavalarias        Wednesday, 20 April 2022   \n",
       "19                     Prof. Laszlo Barabasi        Wednesday, 09 March 2022   \n",
       "20                     Prof. Arthur Spirling     Wednesday, 23 February 2022   \n",
       "21                      Dr. Alexandra Scacco     Wednesday, 09 February 2022   \n",
       "22                    Prof. Margaret Roberts      Wednesday, 26 January 2022   \n",
       "23                        Prof. Lauren Klein      Wednesday, 12 January 2022   \n",
       "24                    Prof. Cesar A. Hidalgo      Wednesday, 1 December 2021   \n",
       "25                   Prof. Christopher Lucas      Wednesday, 3 November 2021   \n",
       "26                        Prof. Camille Roth      Wednesday, 20 October 2021   \n",
       "27                    Prof. Michelle Torrest      Wednesday, 06 October 2021   \n",
       "28                           Prof. Suzy Moat   Wednesday, 22 September, 2021   \n",
       "29                   Prof. Macartan Humphreys   Wednesday, 08 September 2021   \n",
       "30                          Dr. Laura Bronner        Wednesday, 02 June 2021   \n",
       "31                                 Chris Bail        Wednesday, 19 May, 2021   \n",
       "32                         Dr. Melissa Sands          Wednesday, 05 May 2021   \n",
       "33                          Dr. Preslav Nakov       Wednesday, 21 April 2021   \n",
       "34                       Prof. Christo Wilson      Wednesday, 07 April, 2021   \n",
       "35                          John Burn-Murdoch       Wednesday, 24 March 2021   \n",
       "36                               Roman Rivera       Wednesday, 17 March 2021   \n",
       "37                                        NaN   Wednesday, 24 February, 2021   \n",
       "\n",
       "                                                 link  \n",
       "0   https://socialdatascience.network/fall2023/ses...  \n",
       "1   https://socialdatascience.network/fall2023/ses...  \n",
       "2   https://socialdatascience.network/fall2023/ses...  \n",
       "3   https://socialdatascience.network/fall2023/ses...  \n",
       "4   https://socialdatascience.network/spring2023/s...  \n",
       "5   https://socialdatascience.network/spring2023/s...  \n",
       "6   https://socialdatascience.network/spring2023/s...  \n",
       "7   https://socialdatascience.network/spring2023/s...  \n",
       "8   https://socialdatascience.network/spring2023/s...  \n",
       "9   https://socialdatascience.network/polarisation...  \n",
       "10  https://socialdatascience.network/spring2023/s...  \n",
       "11  https://socialdatascience.network/spring2023/s...  \n",
       "12  https://socialdatascience.network/fall2022/ses...  \n",
       "13  https://socialdatascience.network/fall2022/ses...  \n",
       "14  https://socialdatascience.network/spring2022/s...  \n",
       "15  https://socialdatascience.network/spring2022/s...  \n",
       "16  https://socialdatascience.network/spring2022/s...  \n",
       "17  https://socialdatascience.network/spring2022/s...  \n",
       "18  https://socialdatascience.network/spring2022/s...  \n",
       "19  https://socialdatascience.network/spring2022/s...  \n",
       "20  https://socialdatascience.network/spring2022/s...  \n",
       "21  https://socialdatascience.network/spring2022/s...  \n",
       "22  https://socialdatascience.network/spring2022/s...  \n",
       "23  https://socialdatascience.network/spring2022/s...  \n",
       "24  https://socialdatascience.network/fall2021/ses...  \n",
       "25  https://socialdatascience.network/fall2021/ses...  \n",
       "26  https://socialdatascience.network/fall2021/ses...  \n",
       "27  https://socialdatascience.network/fall2021/ses...  \n",
       "28       https://socialdatascience.network/sess9.html  \n",
       "29  https://socialdatascience.network/fall2021/ses...  \n",
       "30       https://socialdatascience.network/sess7.html  \n",
       "31       https://socialdatascience.network/sess6.html  \n",
       "32       https://socialdatascience.network/sess5.html  \n",
       "33       https://socialdatascience.network/sess4.html  \n",
       "34       https://socialdatascience.network/sess3.html  \n",
       "35       https://socialdatascience.network/sess2.html  \n",
       "36       https://socialdatascience.network/sess1.html  \n",
       "37      https://socialdatascience.network/launch.html  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check if length of lists are the same\n",
    "if len(titles) == len(speakers) == len(dates) == len(links):\n",
    "    print(\"All lists are same length\")\n",
    "else:\n",
    "    print(\"Lists are not same length, please ammend\")\n",
    "\n",
    "# Specify the file path including the data folder to save the schedule.csv file\n",
    "schedule_csv_file_path = os.path.join(data_folder_path, 'schedule.csv')\n",
    "\n",
    "# Use the `pd.DataFrame()` function to create a data frame + Save pandas data frame df to a CSV file.\n",
    "df_schedule = pd.DataFrame({'title': titles, 'speaker': speakers, 'date': dates, 'link': links})\n",
    "df_schedule.to_csv(schedule_csv_file_path, index=False)\n",
    "\n",
    "# Double-check that the CSV file was created correctly by opening it using pandas\n",
    "df_schedule = pd.read_csv(schedule_csv_file_path)\n",
    "df_schedule"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "## 📆 Part 2: Scraping for talk links, titles, speakers and descriptions from CIVICA agendas\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scrape the **titles, speakers and descriptions of talks** and save them to a list (using df_schedule as reference)\n",
    "\n",
    "The agenda/talks for each event are found on the individual event page. A `for` loop is used to go through each event link and scrape the relevant data\n",
    "\n",
    "<u>Talk Titles</u>\n",
    "+ All individual talk titles are represented within a `<h4>` tag under a `<div class=\"row schedule-item\">` (found by inspecting the event page)\n",
    "+ However, this also includes the name of the assigned speaker for the talk within a further `<span>` tag (we do not want to include this)\n",
    "+ Therefore: `'div.row.schedule-item h4:not(:has(span))::text'` is a good CSS Selector to use as it included the title of the talk but excludes the name of the speaker\n",
    "\n",
    "\n",
    "<u>Talk Speakers</u>\n",
    "+ All talk speakers are represented within a `<span>` tag under a `<h4>` tag under `<div class=\"row schedule-item\">` (found by inspecting the event page)\n",
    "+ Therefore: `'div.row.schedule-item h4 span ::text'` is a good CSS Selector to use as it includes the name of the speaker but excludes other information under the `<h4>` tag like title of the talk\n",
    "\n",
    "\n",
    "<u>Talk Descriptions</u>\n",
    "+ All talk descriptions are represented within a `<p>` tag under `<div class=\"row schedule-item\">` (found by inspecting the event page)\n",
    "+ Therefore: `'div.row.schedule-item p'` is a good CSS Selector to use"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "talk_title = []\n",
    "talk_speaker = []\n",
    "talk_description = []\n",
    "\n",
    "# For loop to go through each event link using the existing event links column from Part 1 df_schedule dataframe\n",
    "for i in range(len(links)):\n",
    "    url = links[i]\n",
    "    response = requests.get(url)\n",
    "    sel = Selector(text=response.text)\n",
    "\n",
    "    # Getting all the talk titles and excluding the names of the talk speakers\n",
    "    event_agenda = sel.css('div.row.schedule-item h4:not(:has(span))::text').getall()\n",
    "    talk_title.append(event_agenda)\n",
    "\n",
    "    # Getting all the talk speakers\n",
    "    event_speaker = sel.css('div.row.schedule-item h4 span ::text').getall()\n",
    "    talk_speaker.append(event_speaker)\n",
    "\n",
    "    # Getting all the talk descriptions\n",
    "    event_description = sel.css('div.row.schedule-item p').getall()\n",
    "    talk_description.append(event_description)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:70%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "The only thing lending this solution down is that this for loop could have been a function!\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Flattening/Ammending lists for links, titles, speakers and descriptions of talks \n",
    "\n",
    "- In order to use the `pd.DataFrame()` function to create a data frame all the lists/arrays must be the same length\n",
    "- The below code goes through each nested list and either flattens it out or ammends it such that all 4 lists are the same length\n",
    "- In the case of final_talk_description, we also clean up the data to remove unecessary tags"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### List for Titles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Turning the nested title list into a flat list (same will be done with other lists to ensure everything is same length)\n",
    "final_talk_title = [item for sublist in talk_title for item in sublist]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:70%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "Great use of list comprehension to 'flatten out' a nested list\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### List for Speakers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Ensuring the final speaker list corresponds with the final talk titles list\n",
    "final_talk_speaker = []\n",
    "for counter in range(len(talk_title)):\n",
    "    for i in range(len(talk_title[counter])):\n",
    "        if i < len(talk_speaker[counter]):\n",
    "            final_talk_speaker.append(talk_speaker[counter][i])\n",
    "        #If the talk doesn't have a speaker then this will be left blank eg. most 'Announcement' talks do not have an assigned speaker\n",
    "        else:\n",
    "            final_talk_speaker.append('')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### List for Descriptions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# If 'div.row.schedule-item p ::text' was used then the description was split into separate descriptions if there was additional text inside <b>\n",
    "# To avoid this problem, '::text' was left out, and instead the now included <p> and <b> tags were removed using a for loop\n",
    "for i in range(len(talk_description)):\n",
    "    for j in range(len(talk_description[i])):\n",
    "        # Remove <p> and <b> tags using string.replace()\n",
    "        talk_description[i][j] = talk_description[i][j].replace(\"<p>\", \"\").replace(\"</p>\", \"\").replace(\"<b>\", \"\").replace(\"</b>\", \"\").replace(\"amp;\", \"\")\n",
    "# Turning the nested list into a flat list (same will be done with other lists to ensure everything is same length)\n",
    "final_talk_descriptions = [item for sublist in talk_description for item in sublist]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### List for Links"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Currently the list of links will not match the parallel list for individual talks, so this is ammended below\n",
    "final_links = []\n",
    "for counter in range(len(talk_title)):\n",
    "    for i in range(len(talk_title[counter])):\n",
    "        final_links.append(links[counter])            "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Convert the lists to a **pandas data frame** and save it to a **CSV file**\n",
    "\n",
    "- Note: In order to use the `pd.DataFrame()` function to create a data frame all the lists/arrays must be the same length. This was checked using `if` statement\n",
    "- The final CSV file (agenda.csv) is saved directly to the data folder creating in the Setting Up stage\n",
    "- The final CSV file (agenda.csv) can be viewed as a table\n",
    "- If an event has no speakers, then the relevant cell of the table will be left empty"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All lists are same length\n"
     ]
    },
    {
     "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>event_link</th>\n",
       "      <th>talk_title</th>\n",
       "      <th>talk_speakers</th>\n",
       "      <th>talk_description</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>https://socialdatascience.network/fall2023/ses...</td>\n",
       "      <td>Welcome Introduction</td>\n",
       "      <td>Dr. Ghita Berrada, LSE</td>\n",
       "      <td>Setting the scene: Brief intro to the speaker ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>https://socialdatascience.network/fall2023/ses...</td>\n",
       "      <td>Seminar Session</td>\n",
       "      <td>Dr. Divya Srivastava, LSE</td>\n",
       "      <td>Promoting the systematic use of real-world dat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>https://socialdatascience.network/fall2023/ses...</td>\n",
       "      <td>Research Discussion.</td>\n",
       "      <td>Lead Institution</td>\n",
       "      <td>Q&amp;A / Discussion on the research</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>https://socialdatascience.network/fall2023/ses...</td>\n",
       "      <td>Announcement</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Upcoming seminar in the series and other annou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>https://socialdatascience.network/fall2023/ses...</td>\n",
       "      <td>Welcome Introduction</td>\n",
       "      <td>Prof. Petra Novak, CEU</td>\n",
       "      <td>Setting the scene: Brief intro to the speaker ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>https://socialdatascience.network/sess2.html</td>\n",
       "      <td>Announcement</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Upcoming seminar in the series and other annou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147</th>\n",
       "      <td>https://socialdatascience.network/launch.html</td>\n",
       "      <td>Welcome Introduction</td>\n",
       "      <td>Prof. Slava Jankin, Hertie School Data Science...</td>\n",
       "      <td>Setting the scene: Context and Goals for the C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>https://socialdatascience.network/launch.html</td>\n",
       "      <td>Institutional Update</td>\n",
       "      <td>Chair: Prof. Kenneth Benoit, LSE Data Science ...</td>\n",
       "      <td>Introducing CIVICA Partner Institutions' Direc...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149</th>\n",
       "      <td>https://socialdatascience.network/launch.html</td>\n",
       "      <td>Round Table Discussion</td>\n",
       "      <td>Dr. Erica Thompson, LSE Data Science Institute</td>\n",
       "      <td>Topic: Data Science and Digital Transformation...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>150</th>\n",
       "      <td>https://socialdatascience.network/launch.html</td>\n",
       "      <td>Closing</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Closing remarks</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>151 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            event_link  \\\n",
       "0    https://socialdatascience.network/fall2023/ses...   \n",
       "1    https://socialdatascience.network/fall2023/ses...   \n",
       "2    https://socialdatascience.network/fall2023/ses...   \n",
       "3    https://socialdatascience.network/fall2023/ses...   \n",
       "4    https://socialdatascience.network/fall2023/ses...   \n",
       "..                                                 ...   \n",
       "146       https://socialdatascience.network/sess2.html   \n",
       "147      https://socialdatascience.network/launch.html   \n",
       "148      https://socialdatascience.network/launch.html   \n",
       "149      https://socialdatascience.network/launch.html   \n",
       "150      https://socialdatascience.network/launch.html   \n",
       "\n",
       "                  talk_title  \\\n",
       "0      Welcome Introduction    \n",
       "1           Seminar Session    \n",
       "2      Research Discussion.    \n",
       "3              Announcement    \n",
       "4      Welcome Introduction    \n",
       "..                       ...   \n",
       "146            Announcement    \n",
       "147    Welcome Introduction    \n",
       "148     Institutional Update   \n",
       "149  Round Table Discussion    \n",
       "150                 Closing    \n",
       "\n",
       "                                         talk_speakers  \\\n",
       "0                               Dr. Ghita Berrada, LSE   \n",
       "1                            Dr. Divya Srivastava, LSE   \n",
       "2                                     Lead Institution   \n",
       "3                                                  NaN   \n",
       "4                               Prof. Petra Novak, CEU   \n",
       "..                                                 ...   \n",
       "146                                                NaN   \n",
       "147  Prof. Slava Jankin, Hertie School Data Science...   \n",
       "148  Chair: Prof. Kenneth Benoit, LSE Data Science ...   \n",
       "149     Dr. Erica Thompson, LSE Data Science Institute   \n",
       "150                                                NaN   \n",
       "\n",
       "                                      talk_description  \n",
       "0    Setting the scene: Brief intro to the speaker ...  \n",
       "1    Promoting the systematic use of real-world dat...  \n",
       "2                    Q&A / Discussion on the research   \n",
       "3    Upcoming seminar in the series and other annou...  \n",
       "4    Setting the scene: Brief intro to the speaker ...  \n",
       "..                                                 ...  \n",
       "146  Upcoming seminar in the series and other annou...  \n",
       "147  Setting the scene: Context and Goals for the C...  \n",
       "148  Introducing CIVICA Partner Institutions' Direc...  \n",
       "149  Topic: Data Science and Digital Transformation...  \n",
       "150                                    Closing remarks  \n",
       "\n",
       "[151 rows x 4 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check if length of lists are the same\n",
    "if len(final_links) == len(final_talk_descriptions) == len(final_talk_speaker) == len(final_talk_title):\n",
    "    print(\"All lists are same length\")\n",
    "else:\n",
    "    print(\"Lists are not same length, please ammend\")\n",
    "\n",
    "# Specify the file path including the data folder to save the agenda.csv file\n",
    "csv_file_path = os.path.join(data_folder_path, 'agenda.csv')\n",
    "\n",
    "# Use the `pd.DataFrame()` function to create a data frame + Save pandas data frame df to a CSV file.\n",
    "df_agenda = pd.DataFrame({'event_link': final_links, 'talk_title': final_talk_title, 'talk_speakers': final_talk_speaker, 'talk_description': final_talk_descriptions})\n",
    "df_agenda.to_csv(csv_file_path, index=False)\n",
    "\n",
    "# Double-check that the CSV file was created correctly by opening it using pandas\n",
    "df_agenda = pd.read_csv(csv_file_path)\n",
    "df_agenda"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:70%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "The remaining of this notebook is very impressive and 🆒. It shows initiative, creativity and it is insightful. It is also very well documented, we know exactly what is going on.\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "## 💡 Additional Insights into this information"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### I will be looking at the following insights about the data from the CIVICA Seminar Website\n",
    "\n",
    "**1. 📖 Natural Language Analysis of Topics of Events (to determine what kinds of events are most common)**\n",
    "\n",
    "**2. 🗣️ Frequency of different speakers**\n",
    "\n",
    "**3. 📈 Frequency of CIVICA seminar events per quarter**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "### 📖 Natural Language Toolkit and Analysis of Most Popular Topics for CIVICA Data Science Seminar Events\n",
    "\n",
    "- I thought that it would be very interesting to see what the most popular topics/points of dicussion for events and talks were\n",
    "- I realised that I could explore this via NLTK (Natural language toolkit)\n",
    "- 'Stop words' like 'The', 'And', 'To' etc. will be filtered out, leaving us only with words that could potentially give us an indication into the most popular topics for events to be held about. Some custom words were added to the list of 'Stop Words' if they appeared in every description eg. \"Q&A, and the first few words were not included in the graph as they consisted of very common words like \"Data\" which gives us little new insight (given that this is a seminar series about data, obviously Data will be the most common word)\n",
    "- The most popular words in event titles and talk descriptions will be outputted in the form of a bar chart for easy visual digestion\n",
    "- Key insights and takeaways from the output are included below the bar chart"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to /Users/jon/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "import certifi\n",
    "import os\n",
    "\n",
    "# Set the SSL_CERT_FILE environment variable to the path of the updated certificates bundle\n",
    "os.environ['SSL_CERT_FILE'] = certifi.where()\n",
    "# Download the stopwords dataset (only required once)\n",
    "nltk.download('stopwords')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "\n",
    "# List of words\n",
    "title_words_list = [word for titles in titles for word in titles.split()]\n",
    "description_words_list = [word for final_talk_descriptions in final_talk_descriptions for word in final_talk_descriptions.split()]\n",
    "\n",
    "event_overview = []\n",
    "# For loop to go through each event link using the existing event links column from Part 1 df_schedule dataframe\n",
    "for i in range(len(links)):\n",
    "    url = links[i]\n",
    "    response = requests.get(url)\n",
    "    sel = Selector(text=response.text)\n",
    "\n",
    "    # Getting all the talk titles and excluding the names of the talk speakers\n",
    "    overview = sel.css('div.col-lg-9.fade.show.active p::text').getall()\n",
    "    event_overview.extend(overview)\n",
    "\n",
    "\n",
    "event_overview_list = [word for event_overview in event_overview for word in event_overview.split()]\n",
    "\n",
    "combined_words_list = title_words_list + description_words_list + event_overview_list\n",
    "\n",
    "# Get the list of English stopwords from nltk\n",
    "stop_words = set(stopwords.words('english'))\n",
    "\n",
    "# Additional words to be added to stop_words\n",
    "custom_stop_words = ['us', 'Setting', 'scene:', 'University', 'Professor', 'also', 'session', 'policy.', 'social,', 'ten', 'Hertie', 'Series,', 'sciences,', 'Hertie', 'School', 'Professor', '.', 'political,']  # Add your custom stop words here\n",
    "\n",
    "# Extend the stop_words set with custom_stop_words\n",
    "stop_words.update(custom_stop_words)\n",
    "\n",
    "# Remove stopwords from the list of words\n",
    "filtered_words_list = [word for word in combined_words_list if word.lower() not in stop_words]\n",
    "\n",
    "# Count the filtered words\n",
    "filtered_counted_words = Counter(filtered_words_list)\n",
    "\n",
    "items_list = []\n",
    "frequencies_list = []\n",
    "\n",
    "# Iterate through counted items and frequencies\n",
    "for item, frequency in filtered_counted_words.most_common()[18:46]:\n",
    "    items_list.append(item)\n",
    "    frequencies_list.append(frequency)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Sample data\n",
    "items_list\n",
    "frequencies_list\n",
    "\n",
    "# Create a bar chart\n",
    "plt.bar(items_list, frequencies_list)\n",
    "\n",
    "# Adding labels and title\n",
    "plt.xlabel('Key Word')\n",
    "plt.ylabel('Frequency')\n",
    "plt.title('Most Popular Keywords in CIVICA Data Science Seminar Events and Talk Descriptions')\n",
    "\n",
    "# Display the frequency of each bar above the bars\n",
    "for i, freq in enumerate(frequencies_list):\n",
    "    plt.text(i, freq + 0.5, str(freq), ha='center')\n",
    "\n",
    "# Rotate x-axis labels sideways\n",
    "plt.xticks(rotation='vertical')\n",
    "\n",
    "# Display the graph\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 📖 Takeaways from Natural Language Analysis of Topics and Descriptions\n",
    "\n",
    "- From the bar chart we can see that the seminars tend to focus a lot on uses of AI/Data Science (high frequency of \"applications\", \"methodologies\", \"use\")\n",
    "- There is a heavy emphasis on the \"social\", \"public\" and \"economic\" aspects on uses of AI and Data Science\n",
    "- The lack of technical jargon in the event descriptions and titles shows that these talks will be accessible to the general public/to audiences that do not have existing knowledge of data science concepts (This is a good thing in my opinion! As it means that the CIVICA events are accessible to a larger group of people!)\n",
    "- The fact that \"World\" appears more times than \"European\" might indicate that despite CIVICA being a european based organisation, their world is applicable to a broader scope/to the entire world (not just focussing on europe)\n",
    "- The high frequency of worlds like \"multi-disciplinary\", \"unique\", \"leading\" and \"new\" potentially indicates that the CIVICA seminar series is good at staying ahead of the data science curve, and that we have been successful at bringing in speakers that are at the cutting edge of the field (or maybe the descriptions are just written in such a way to bring in more interest to the events)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "### 🗣️ Seeing who the most common event speakers and talk speakers are across all events\n",
    "\n",
    "- I thought it would be interesting to see who the most frequent speakers at all the CIVICA events are (as both main speakers and assigned speakers for individual talks within events)\n",
    "- If there are any speakers who have spoken a lot, then CIVICA might make note to not bring them in for future events and to maybe bring in a different speaker to allow for more variety (alternatively, they might also want to give an acknowledgement to any speaker that has spoken lots at events!)\n",
    "- A bar chart has been generated to visualise this information, and the code has been written such that events with no speaker will be shown as \"no assigned speaker\" as opposed to just showing a blank space"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "\n",
    "# List of all speakers\n",
    "all_speakers = final_talk_speaker + speakers\n",
    "counted_speakers = Counter(all_speakers)\n",
    "\n",
    "speaker_list = []\n",
    "occurence_list = []\n",
    "\n",
    "# Iterate through counted items and frequencies\n",
    "for item, frequency in counted_speakers.most_common(10):\n",
    "    if item==\"\":\n",
    "        speaker_list.append(\"No Assigned Speaker\")\n",
    "        occurence_list.append(frequency)\n",
    "    else:\n",
    "        speaker_list.append(item)\n",
    "        occurence_list.append(frequency)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Sample data\n",
    "speaker_list\n",
    "occurence_list\n",
    "\n",
    "# Create a bar chart\n",
    "plt.bar(speaker_list, occurence_list)\n",
    "\n",
    "# Adding labels and title\n",
    "plt.xlabel('Speaker Name')\n",
    "plt.ylabel('Number of events/talks spoken at')\n",
    "plt.title('10 Most Frequent Speakers in CIVICA Data Science Seminar Events and Talks')\n",
    "\n",
    "# Display the frequency of each bar above the bars\n",
    "for i, freq in enumerate(occurence_list):\n",
    "    plt.text(i, freq + 0.5, str(freq), ha='center')\n",
    "\n",
    "# Rotate x-axis labels sideways\n",
    "plt.xticks(rotation='vertical')\n",
    "\n",
    "# Display the graph\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 🗣️ Takeaways from analysis of frequency of speakers\n",
    "\n",
    "- From the bar chart we can see that many talks have no assigned speaker or are broadly facilitated by the lead institution\n",
    "- There are no single speakers that have spoken at an incredibly high number of events or talks, indicating that CIVICA seminar series is good at bringing in a variety of speakers (which is good for marketing, as we are always doing new things and not relying on a select froup of speakers)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "### 📈 Seeing how the frequency of events has moved over time\n",
    "\n",
    "- I thought it would be interesting to get an overview of how many CIVICA seminar events are held each quarter to see how consistent these events are\n",
    "- eg. are we gradually holding more or less events, is there significant fluctuation? etc\n",
    "- I used datetime to achieve the final graph, and used nested loops to count the dates into the correct quarters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "\n",
    "# Sample list of dates (in string format)\n",
    "formatted_dates = [string.replace(\",\", \"\") for string in dates]\n",
    "\n",
    "# Initialize quarter and count lists\n",
    "quarters = []\n",
    "counts = []\n",
    "\n",
    "# Initialize quarter and count dictionary\n",
    "quarter_counts = {}\n",
    "\n",
    "# Iterate through the dates and count quarters for each year\n",
    "for date_str in formatted_dates:\n",
    "    # Convert string to datetime object using the appropriate format\n",
    "    date_obj = datetime.strptime(date_str, \"%A %d %B %Y\")\n",
    "    year = date_obj.year\n",
    "    # Filter dates for 2021, 2022, and 2023\n",
    "    if 2021 <= year <= 2023:\n",
    "        quarter = (date_obj.month - 1) // 3 + 1\n",
    "        # Update quarter count in the dictionary\n",
    "        key = f\"Q{quarter} {year}\"\n",
    "        quarter_counts[key] = quarter_counts.get(key, 0) + 1\n",
    "\n",
    "# Extract quarters and counts into separate lists\n",
    "for quarter, count in quarter_counts.items():\n",
    "    quarters.append(quarter)\n",
    "    counts.append(count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Sample data\n",
    "x = quarters\n",
    "y = counts\n",
    "\n",
    "# Reverse the x and y list to display the axis in reverse order\n",
    "x.reverse()\n",
    "y.reverse()\n",
    "\n",
    "# Create a line graph\n",
    "plt.plot(x, y, marker='o', color='b', linestyle='-', linewidth=2, markersize=8)\n",
    "plt.xticks(rotation='vertical')\n",
    "\n",
    "# Adding labels and title\n",
    "plt.xlabel('Date (Quarter, Year)')\n",
    "plt.ylabel('Number of CIVICA Seminar Events')\n",
    "plt.title('Frequency of CIVICA Seminar Talks')\n",
    "\n",
    "# Display the graph\n",
    "plt.grid(True)  # Add gridlines for better readability\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:70%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "There is just one important mistake to the plot above: the Y-axis doesn't start at 0.\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 📈 Takeaways from analysing the frequency of CIVICA seminar events\n",
    "\n",
    "- As we can see from the graph, there is seasonal fluctuation in how many CIVICA Seminar events are held each quarter\n",
    "- Q3 consistently sees the lowest number of seminars (probably due to the fact that this time period contains the summer holidays)\n",
    "- There is no clearly observable trend, but compared to 2021 and 2022, thus far 2023 is falling behind in terms of number of events (even if we account for the fact that there are 2 months left)\n",
    "- This might suggest that in order to hold a similar amount of talks compared to previous years CIVICA should try and schedule in more talks for december 2023\n",
    "- The average decline in number of events may also represent the shift in preference away from online events (like the CIVICA seminar series) to in person events post pandemic. If we can get some additional information on this, then CIVICA may decide to hold more in person events in order to gain more attraction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--- \n",
    "## 💻 In what ways did I use Generative AI tools to help with my assignment?\n",
    "\n",
    "- While working on my assignment, I found Generative AI tools (ChatGPT) to be incredibly useful for debugging specific sections of my code\n",
    "- It helped me identify syntax errors and provided valuable suggestions for correcting loop structures\n",
    "- However, I didn't heavily rely on Generative AI tools for the overall assignment\n",
    "- This was mainly because the responses they generated were often generic and lacked the specificity required for the summative\n",
    "- Additionally, there were instances where the suggestions provided contained errors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"width:70%;border: 1px solid #aaa; border-radius:1em; padding: 1em; margin: 1em 0;\">\n",
    "\n",
    "**(Jon's comments)**\n",
    "\n",
    "This is good and in times very specific (used it for debugging and to correct loops) but I feel like some parts of this notebook - the use of Collections and ordered_set packages - probably came as the output of generative AI. If true, the author failed to mention this.\n",
    "\n",
    "</div>"
   ]
  }
 ],
 "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": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}