AI for Impact Assessment
Build NLP models to classify customer product descriptions to estimate their impact on the environment.
We’re making it possible for companies to take responsibility for their impact, together with Metabolic
Metabolic is a group of organizations that aims to bring the economy within Planetary Boundaries as soon as possible. Recently they have launched a new product, Metabolic Link. It helps companies proactively assess their nature impacts and risks.
This SaaS solution helps companies assess and report their impact on nature and biodiversity, in addition to greenhouse gas (GHG) emissions. It minimizes the problem of data management by basing its analysis on existing GHG data and providing guidance on obtaining additional data. It also uses geospatial data to identify context-specific risks and opportunities and generate detailed insights on where and how to act successfully. Because the impact of water and land use is not the same everywhere on Earth. 🌎
The Challenge we’ll be tackling is that Metabolic receives a lot of data from companies about, for example, procurement and travel expenses in varying formats. Manually parsing what is actually described in that data is a lot of work (and not always possible), so we will automate (and estimate) this with machine learning models, the famous GPT, among others!
In this challenge, we will
1. Do thorough exploratory data analysis to understand the data and problem at hand
2. Enrich the dataset with new features
3. Build a data pipeline that takes any input columns and converts them into a standardized data model
4. Use large language models to perform classification
5. Try out ensemble methods to find the best results
Technologies we will use
- Combinatorial optimization
- Word vectors
Who are we looking for?
We are looking for data science & AI engineers, previous work with NLP would be of great benefit.
You will collaborate with a diverse team of up to 50 international participants in subteams. You can join as a contributor (8-12 hours per week commitment for 10 weeks) or coach (2-4 hours per week, only for experienced ML professionals)
We’ll organize masterclasses on environmental impact assessment and NLP during the challenge to bring you up to speed.
Did you know
🇺🇸 The first country to give importance to Environmental Impact Assessment through its National Environmental Protection Act (NEPA) of 1969 was the United States. A large number of countries followed, which were having industries. Canada, Australia, the Netherlands, and Japan adopted EIA legislation in 1973, 1974, 1981, and 1984, respectively. In July 1985, the European Community (EC) made EIA mandatory.
Our challenge partners
Application Deadline: 11 June 2023
Challenge Kick-off: 12 June 2023
Midterm Presentations: 10 July 2023
Final Presentations: 21 August 2023