This repo reproduces parts of the analysis in the open access paper titled The Nobel family, by Richard S. J. Tol.
The aim of this repo is twofold:
- Demonstrate how to use Kùzu to construct the graph and how to use NetworkX to analyze the data in it
- Enrich the graph with official metadata from the nobelprize.org API. Because the names in either dataset do not match exactly, entity resolution is needed -- this project uses the Senzing API to resolve entities so that the data from these two sources can be effectively merged to analyze the Nobel Laureate network.
The primary dataset for this project includes the tree structures that represents Nobel Laureates between 1901 and 2022 and their mentors (dating back hundreds of years), generously made available under the MIT license by the original author (Richard S. J. Tol). The original data is available within MATLAB files in this repo, but is cleaned up and consolidated for ease of reproducibility in this repo.
The secondary dataset that supplements the Nobel Laureate metadata is obtained from the nobelprize.org API, available freely under a Creative Commons Zero (CC0) license. If you use the data in this repo, please cite the original source (nobelprize.org) as per their license terms.
Install the dependencies via requirements.txt
.
pip install -r requirements.txt
The Kùzu graph is constructed using the script build_kuzu_graph.py
.
$ python build_kuzu_graph.py
Created node table for Physics nobel prize
Created node table for Chemistry nobel prize
Created node table for Medicine nobel prize
Created node table for Economics nobel prize
Inserted scholars into the Scholar node table
Inserted relationships into the MENTORED relationship table
Inserted data into the WON_Scholar_Physics relationship table group
Inserted data into the WON_Scholar_Chemistry relationship table group
Inserted data into the WON_Scholar_Medicine relationship table group
Inserted data into the WON_Scholar_Economics relationship table group
The graph of Nobel laureates and their mentor relationships looks like this:
In total, there are 3,517 scholars who mentored modern Nobel laureates that date back hundreds of years!