The Muller Investigation

banner-logo


Connections and Patterns in the Muller Investigation was developed by Fathom using a tool called Porfiry that deconstructs documents using algorithms and finalizes these results with custom curation tools. The documents are used as a collection of data points. The data extracted from court filings and news articles about the Muller’s investigation in the Trump-Russia connection to create chronological flows for each participant. It reveals connections between this data that would not have been obvious otherwise.


The Data

gtrend_logo


The data comes from the Muller Investigation court filings and any news articles from The New York Times. Custom tools were used to curate machine learning outputs from this data.

Each data point is where a person of interest is mentioned and the approximate date that the article references. This gives a timespan for all of the information rather than just the article publish dates.

How the plots were built


Data Representation

As seen in the animation above, the plots are very dense with data.

  • Each circle represents 1+ articles about that person and 0+ about other people connected to that person in any of the articles
  • A solid circle represents connections to other people
  • An empty circle repreents only connections to that person
  • They are ordered by date reference of the event not the date of the article publishes
  • All of the data points are then places as a row for each person that spans across a timeline in years
  • The scale is approximately from 1987 – February 2019
  • Trends can be seen with the largest circles and proximity of circles
  • The stars at the top tell you significant dates relevant to the data

Explore


The focus of the visualization is on highlighting connections between participants that would not have been obvious otherwise. Each data point can be selected which highlights connections with a line between participants referenced from the data point. The articles that reference these people can be seen which highlights the areas where their names are mentioned. The links to these articles are also attached and you can then select on them to read further.

Colorblind Friendliness

Snow
Forest
Mountains
Anomalous Trichromacy: Red-Weak/Protanomaly
Anomalous Trichromacy: Green-Weark/Deuteranomaly
Anomalous Trichromacy: Blue-Weak/Tritanomaly

Snow
Forest
Mountains
Dichromatic View: Green-Blind/Deuteranopia
Monochromatic View: Blue Cone Monochromacy
Monochromatic View: Monochromacy/Achromatopsia

Since the data comparisons are most obvious by the shapes rather than colors it is easy to see trends. The only color used is green the black and white is very easy to distinguish when trying to see what is or isn’t selected.

What Works...

Snow
Forest


If you want to look deeper into the connections made by the Muller investigation and which people are connected and how you can select the people of interest and see what articles connect them.

  • Great for visualizing connections and then being able to go into depth of these connections.
  • Easily see the chronological timeline of events for each person and where they link in the timeline
  • It is easy to see who is connected to the person featured as they are highlighted green.
  • The people with the most connections to the featured person move to the top of the list. You don't have to scroll through people who aren't connected to find them

What Doesn't Work...

cosmo

Someone’s name might not specifically be mentioned in an article however doesn’t mean they weren’t involved. You can search manually through and “find” and event but there’s no easy way to do this since its mapped by years it’s difficult to find the month and date of where it is. You are only able to see events that are predefined at the top. You also cannot see who else is connected if they are father down. You can't select more than one person to see their connections. For example if you wanted to see Putin's and Trump's connections you can't select on both.

Improvements

cosmo

All of their data comes from the New York Times and the Muller Reports. The NYT is highly factual and one of the most reliable sources in terms of news. However, since it is the only source other than the Muller reports things you want to see may not be in there. Additionally, adding the ability to search for certain events would be helpful when trying to find select pieces of information. On a regular screen you can’t see where the data points go which is not very user friendly since you have to scroll down.

About

Author: Casey Charlesworth

Class: CS424 - Data Visualization