What Billionaires are Alike?

Graham Albers
INST414: Data Science Techniques
3 min readDec 4, 2023
Jeff Bezos (Amazon Founder) at a conference giving a presentation.

Introduction:

Ever wondered about the commonalities among billionaires like Jeff Bezos? Exploring the traits shared by these high achievers can be intriguing, especially in times of crisis. During the devastating wildfires that destroyed the Maui community in Hawaii, calls for assistance from billionaires, including Bezos, echoed. Looking at responses shows patterns that could guide Americans in identifying potential benefactors during future natural disasters. By understanding which billionaires exhibit similarities to Jeff Bezos, a plan emerges for effective use of resources and aid. This insight becomes crucial for creating a successful approach to disaster relief, ensuring a coordinated response from those with the intention to make a significant impact on affected communities.

Similarity Metrics and Data Source:

This data was discovered from kaggle.com the dataset featured information on every billionaire in the Forbes Exclusive Billionaires Club. The data had over 2640 different billionaires or their families. Some of the information featured was net worth, age, name, rank, country, source of wealth, education, and industry. From this, it was determined that age, net worth, and education, specifically a bachelor’s degree, would be used as the main features to compare the different billionaires to Jeff Bezos. Specifically, euclidean distance will be used to compare these elements along with cosine similarity, and dimensionality reduction.

Similarity Results:

Analyzing billionaire profiles based on net worth, education, and age, similarity metrics — euclidean distance, cosine similarity, and dimensionality reduction — were used. The goal was to identify the ten billionaires most like Jeff Bezos.

Similarity using Euclidean Distance
Similarity using Cosine Similarity
Similarity using Dimensionality Reduction PCA
A comparison using Euclidean Distance
A comparison using Cosine Similarity
A comparison using Dimensionality Reduction

Software Used:

The data analysis started with using the Pandas library to convert a CSV file into a structured data frame. After that, the data frame was cleaned since only the net worth, education, name, and age columns were being used. Then sklearn and scipy modules were used to compute different similarity metrics such as Euclidean distance, cosine similarity, and dimensionality reduction. This looked into the relationships between the different billionaires and Jeff Bezos. The Matplotlib library then enabled the creation of bar graphs and made the visualizations for the ten most similar billionaires to Jeff Bezos.

Cleaning The Data:

To clean the data many of the original columns were removed from the original data frame. Some of the columns removed included, country, industry, forbes_id, rank, children, marital status, and others. That left us with only the name columns as the index for the data frame and then the net worth, education, and age columns also remained.

Limitations:

It is very important to note that some of the people on the list were handed their wealth through their bloodline and did not have to work for it. It is also important to note that whole families are in the dataset and occasionally will be included in the statistics. Only using the metrics that were used tried to account for this and also create an even playing field for every billionaire in the dataset. If you would like to view the code you can visit the GitHub repository right here!

--

--