The Privatization of AI Research: Causes and Consequences

We analyze the causes and consequences of the ongoing privatization of AI research, particularly the transition of AI researchers from academia to industry. This is a collaborative work between the Aalborg University and Nesta.

How to use the data / repo

• Clone the repository with

git clone https://github.com/nestauk/ai_research

• Do cd ai_research to change your working directory to the project's repo and run make create_environment. This will create a new Anaconda environment and install all the project dependencies.
• conda activate ai_research to activate the newly created anaconda environment.
• In a jupyter notebook, you can do the following to read a data table:

import pandas as pd
from sqlalchemy import create_engine
from sqlalchemy.orm import sessionmaker
from ai_research.mag.mag_orm import FieldOfStudy

# Read the configuration file and create a session.
db_config = 'postgres+psycopg2://USER:PASSWORD@HOST:5432/DBNAME'
engine = create_engine(db_config)
Session = sessionmaker(engine)
s = Session()

fields_of_study = pd.read_sql(s.query(FieldOfStudy).statement, s.bind)

Note: mag_orm.py contains the SQLAlchemy mappings (ORMs) used in the database. In the example above, I imported the FieldOfStudy ORM which corresponds to mag_fields_of_study table in the database. You have to import the ORMs for the tables you want to read!

Data

Sources:

• Microsoft Academic Graph
• Google Places API

Data decisions

• Timeframe: 2000-2020
• We collect MAG papers containing one of the following Fields of Study:
  • deep learning
  • machine learning
  • reinforcement learning
• We keep document with and without a DOI

Project based on the Nesta cookiecutter data science project template.