“We’re all possibly Frank Sinatra’s son” : Significant Network around Ronan Farrow


Visualization

Introduction

Ronan Farrow is an American journalist as well as former Special Adviser of the Obama administration. It has been known that he was born to actress Mia Farrow and filmmaker Woody Allen, but in 2013, Mia Farrow raised speculation that singer-actor Frank Sinatra was Farrow’s biological father. Ronan joked about this on his Twitter, “Listen, we’re all *possibly* Frank Sinatra’s son.” Mia Farrow was once married to Frank Sinatra in 1966, when she was 21 years old and Sinatra was 50, but they were divorced 2 years later. Mia Farrow said in an interview with Vanity Fair in 2013 that she and Sinatra “never really split up”. Mia Farrow gave birth to Ronan in 1987.

Around Ronan Farrow, I found many complex relationship from Frank Sinatra to Mia Farrow to Woody Allen to Andre Previn, a German-American pianist and conductor. By researching previous articles, Wikipedia, and interviews, I found the personal network around Ronan Farrow, listed them, and decided to visualize them using Gephi.

Data Preparation

1. Data source

Starting with a Wikipedia page about “Personal Life of Frank Sinatra”, I researched interviews and articles about people around Ronan Farrow. I set four main people as main posts of the network: Frank Sinatra, Mia Farrow, Woody Allen (who was a long time partner of Mia Farrow), and Andre Previn, who once a husband of Mia Farrow for the following reasons:

  • Mia Farrow & Frank Sinatra are assumed to be Ronan’s biological parents.
  • Woody Allen was officially known as father of Ronan Farrow. They are estranged from each other since Woody Allen had relationship with Soon-yi Previn, once an adopted child of Mia Farrow and Andre Previn. Ronan was given the last name “Farrow” to avoid a family with “one child named Allen amidst two Farrows and six Previns.”
  • Mia Farrow and Andre Previn were married between 1970 and 1979. Ronan was grown up with siblings from his mothers’ previous marriage.

The people on this network is limited to people who had personal relation with these four people, such as spouses, biological / adopted children, and revealed partners. The network expands to people who were married to or had personal relationship with, and does not consider additional relation with people in children’s generation.

2. Value 

After listed up all people in the network, I assigned values according to the types of relationship:

  • 1- Marital relations
  • 2- Partner, extra marital affair
  • 3- children (both adopted and biological)

3. Export to CSV 

This CSV file contains source, target, value, and additional comments for further reference.

 

Visualization examples

One of the most popular visualization using Gephi is social network visualization. I envisioned my data will be formed similar to social network visualization, therefore I found some examples in social network visualization.

This network graph shows Twitter network around one person. Each tie represents the first person the author mentioned in one of his past 200 tweets.

Below example shows a Facebook friends network. This author shows three different vis trying different measures out of one set of data.

  

At first I consider the first example as my design choice, but the network of my data does not spread out from only one person. So I used ForceAtlas2 layout and tried different measures to find meaningful one.

 

Visualization Using Gephi 

I imported the CSV file into Gephi and chose ForceAtlas2 layout. In all graphs, the colors of the edges show different relationship using partition of the edges.

  • Red – 1- Marital relations
  • Yellow – 2- Partner, extra marital affair
  • Cyan – 3- children (both adopted and biological)

This is vis that I configured degree: the size of the node is proportional to the number of connections from the node.

I also configured closeness centrality and changed color and size of the nodes along with the measure. Closeness centrality of a node is the average length of the shortest path between the node and all other nodes in the graph. In other words, the more central a node is, the closer it is to all other nodes.

Further Discussion

Since I did not find a complete dataset about my topic, I researched and made csv file from scratch. In this dataset I did not specify sibling relationship.

Also this vis does not show time lapse. For example, Mia Farrow and Frank Sinatra are connected in marriage in the vis, but Ronan Farrow was born long after they got divorced. Since this vis does not reflect time differences, some users  may misunderstand some relationship. In the further research I would like to explore if I can show transition of the relationship as time changes.

Reference (Not specified in the text) 

http://www.vanityfair.com/online/daily/2013/10/mia-farrow-children-family-scandal
https://en.wikipedia.org/wiki/Frank_Sinatra
http://www.nydailynews.com/entertainment/gossip/frank-sinatra-love-marriages-article-1.329594
https://en.wikipedia.org/wiki/Centrality#Closeness_centrality
http://www.notablebiographies.com/Sc-St/Sinatra-Frank.html

 

 

 

No related posts.