Facebook Messenger Analysis

Have you ever wondered about all the personal data that Facebook collects on its over 2 billion users? It’s time to harness the information Facebook has on you for your own good and discover some insights. Tune into our livestream this Thursday where we’ll show you how to download and analyze your Facebook messenger chat data and get a better understanding of your relationships with your friends.

In order to follow along with the stream on your own computer and using your own Facebook data, complete the below steps before Thursday evening:

• Install Python and Jupyter Notebooks on your computer with Anaconda / Miniconda (see below for further instructions).
• Download your Facebook Messenger Chat Data
  • Go to www.Facebook.com and click the arrow on the top right to access your “Settings”
  • Click on “Your Facebook Information”
  • Click “Download Your Information”
  • Update the “Format” field to “JSON” and the “Media Quality” field to “Low”
  • Click “Deselect All” on the right-hand side
  • Click the check-box for “Messages”
  • Click “Create File”. Facebook will notify you when the file is ready for download
  • Download your file

On Thursday we will post the link to the livestream here, so make sure to check back then! Get excited for all the juicy insights you soon shall find.

Installation: Miniconda

This video details how to download and install Miniconda.

To install Miniconda, follow these steps:

1. Navigate to the Miniconda download page: Miniconda

2. Select the Python 3.6 installer for your computer’s operating system.

3. Locate the installer that you downloaded using Explorer (Windows) or Finder (Mac OS).

4. Run the installer. Use the following instructions based on your computer’s operating system:

Mac OS:

1. You may receive a notification about XCode requiring additional component. Click “Install” and enter your password to proceed.

2. Open your terminal and navigate to the folder where you downloaded the installer. Type the following command in the terminal and press “Return” on your keyboard:

bash miniconda-filename.sh

miniconda-filename.sh is a fictional file name in the example above. Your file name will look something like Miniconda3-latest-MacOSX-x86_64.sh.

3. Follow all instructions in the terminal (you can press Enter as-needed and type yes when necessary).

Windows:

1. Follow the installation instructions provided by the installer.

Was the Installation Successful?

To test whether your installation was successful (regardless of your computer’s operating system), type the following command into your terminal:

conda list

You should see a list of all the packages that Miniconda installed. If you’re on a computer that uses Windows, you may have to first navigate to the folder where you installed Miniconda for the conda list command to function properly.

Congrats! You now have Miniconda (with Python 3.6) installed on your computer, and you are ready for some data science!

Installing Jupyter

To install Jupyter with Anaconda / Miniconda, enter the following on the command line:

conda install jupyter

You may need to enter y to confirm the install. conda will install Jupyter, an open-source web application that allows you to create and share documents that contain live code, equations, visualizations and narrative text.

To learn more about conda, visit the Conda documentation at the following link:

• Getting Started with Conda.

Required Data Science Packages

To make the most of Anaconda / Miniconda in our analysis, you’ll need the following data science packages. Use conda install to install them.

• pandas
• nltk
• matplotlib

For example:

conda install pandas