Startup Genome Partners on Data Science Education Challenge

Startup Genome partnered with Brazilian Edtech platform Sirius to host a challenge for students enrolled in its Master in Data and Decision Science program. Focused on the intricacies of startup ecosystems, the practical challenge was completed by groups of students as part of the technical and analytical skill training required for the program, and highlighted five competencies: business intelligence and statistical analysis, predicting the future with machine learning and statistics, diagnostics and optimization for decision science, fundamentals of data engineering with Python and SQL, and design of data production and service.

Startup Genome provided a dataset on patents and students were asked to identify factors including the number of patents by ecosystem per year, growth rate, moving average, and number of patents by sub-sector, ultimately ranking each ecosystem under a set criteria.

Parshant Sharma, Data Science Lead at Startup Genome, explains that patents data is very important in evaluating a startup ecosystem, saying “Patents help us to identify the scale of innovation happening in the ecosystem and it’s very interesting to look at this data on a sub-sector level. Analysis at this level helps us to identify new and emerging technologies in a startup ecosystem.”

Three student groups responded to the challenge, and their outcomes can be viewed on Sirius’s YouTube channel, alongside other student projects. Subtitles can be viewed in any language by clicking the gear icon for Settings, turning on subtitles in Portuguese, then selecting auto translate in your desired language.

Interested in how we evaluate startup ecosystems? Learn more in University Funding an Indicator of Startup Success.