NLP / AI AGAINST CARBON IMPACT
Turning transaction text into a carbon signal
The challenge explores how machine learning can help estimate the carbon impact of purchases from bank transactions. Understanding spending patterns is a step toward making more informed sustainability decisions.
01
Read the transaction
Use natural-language processing to infer the type of purchase from transaction descriptions.
02
Find its category
Work with Yayzy’s classification system of about 90 categories and its database of more than 30,000 retailers, estimating categories for unfamiliar merchants.
03
Estimate the impact
Combine the category and transaction amount with a weighting factor to estimate carbon impact and help people understand their consumption patterns.
🎯 GOAL: Build an accurate and robust natural language model that can classify and predict the estimated carbon footprint of a single transaction.
By providing insight into consumer patterns we can help people make more sustainable choices. We will be using NLP to classify the type of transaction, eg. “Food & Dining”, “Shopping”. We use this information to predict the CO2 emissions.
In this challenge, we will
Build language models for classification
Use clustering methods to find new categories
Analyse and enrich large datasets (>500.000 instances)
Integrate models into the Yayzy application
Technologies we will use
NLP
Word2Vec
Clustering/Classification
BERT
Huggingface 🤗
🙌 Results & Final Presentation
Check out our final presentation slides here.
✨ Team (Modeling)
Patryk Neubauer
Noelia Otero
Anete Stine Teimane
Nathanya Queby S.
