A Python-Based Methodology for Solving Sustainability Problems with Data Science
09-24, 10:45–11:20 (Europe/Lisbon), Auditorium

Python, as a one-stop-shop for Machine Learning, Geospatial Analytics, Optimization Algorithms, and Visualization Tools, can be leveraged to create a simple yet effective sustainability decision-making methodology with just four steps: Geospatial Indexing, Feature Engineering, Predictive Scoring, and Score Optimization.

With it, it's possible to tackle distinct sustainable development issues, from mobility transition to light pollution in a way that decision-makers can quickly implement it to take action for the Sustainable Development Goals.


How can we solve sustainability problems through data?

AI, Data Science, and Analytics brought immense advantages to many businesses. However, there seems to be a gap concerning sustainability, in which decision-making is not yet benefitting as much as it could from these technologies.

This talk takes a deep dive into this topic and how Python can help us bridge the gap by being a one-stop shop for all kinds of tools required to approach these problems. We propose a simple 4-step data science methodology: Geospatial Indexing, Feature Engineering, Predictive Scoring, and Score Optimization.

We will dive into the specificities of each one of these steps and go through their requirements, dependencies and applications. Afterward, we will move to the deployment of solutions and how they translate into tangible impact on several issues with four case studies.

This methodology has been successfully used in different Data Science competitions, including the winning solution for the World Data League, which gathers the best data scientists to solve social impact challenges.

22-year-old Portuguese student, finishing my Master's in Industrial Engineering and Management at the University of Porto. With extensive experience in business, strategy, and engineering case study competitions, as well as in data science hackathons, both nationally and internationally.
I've been active in several juvenile associations, most prominently, as President of ShARE-UP, a junior association that develops a training program for students and consulting projects for startups and businesses, focusing on combining economic performance with societal progress.
Professionally, I've worked at Smartex.ai, a fast-growing Portuguese start-up that is disrupting the textile industry by providing AI tools to minimize waste production, and its environmental consequences. Besides, I've used analytical and machine learning tools in a brief internship at Sonae Fashion, as part of the Customer Strategy team.
Currently looking for ways to apply business, analytical and technical concepts to solve social or environmental issues.

Biomedical Engineer working as Data Scientist in a fast-growing med-tech startup. My professional career started in 2017, first collaborating on several research projects and then moving to a multinational IoT company before changing to my current position. My goal is to deliver real impact through analytics, AI and communication, always focused on sustainable development.

MSc in Bioengineering with a specialization in Biomedical Engineering. Currently working as a Data Scientist at Automaise, developing AI-powered solutions to improve customer care.

Portuguese, 22, and master’s student of Biomedical Engineering at the University of Porto. My name is João Matos and I am passionate about traveling, having lived in 4 different countries. I am currently living in Japan and I will move to the US in October to develop my master’s thesis at MIT. As a Biomedical Engineering to-be, I want to bring AI and clinical data together to improve healthcare and people’s quality of life. Driven by delivering a good impact with my analytical and technical skills, I am also interested in solving sustainability and societal issues. I was President of ShARE-UP, a Do Well Do Good international consulting club that promotes social progress. More recently, alongside my teammates, I won a couple of datathons with challenges that tackled some of the United Nations’ Sustainable Development Goals.