Creative Innovation: Data

How do we determine innovation?

Since innovation is an intangible concept, there currently isn't a true indicator or metric for innovation. As such, we will use an unsupervised model where nations with similar levels of innovation will be clustered together.

Additionally, we will be able to use the Global Innovation Index as a benchmark for our supervised learning approach, and compare the results of our two models.
What are potential predictors of innovation?

We intended to analyze a variety of measures (including creative and economic factors) to determine which of these are potentially significant predictors of what makes a nation innovative.
What data was used in our analysis?

We obtained and integrated 40 datasets from various sources such as the World Bank, the Global Innovation Index site, and UNESCO. Some datasets that we looked at include:
  • books published per country per year
  • scientific publications
  • GDP
Our datasets also contained time series data ranging from 1950 to 2020 for a variety of countries.

Many countries also had sparse data, which presented an additional challenge during the pre-processing phases. After normalizing and creating an integrated dataset, we proceeded with our analysis.