MIDS Capstone Projects: Fall 2020

BugTrAP utilizes machine learning models to automate data annotation from images. This entomology data digitization will enable scientists to...
Carzam is a vehicle recognition product that can be an enabler for various industries such as marketing, law enforcement or security. It uses cutting edge...
Communication Unbreakdown helps community organizers reach target populations for civic engagement. Their solution was used by People's Action to find...
DeciBull upscales low quality audio to higher resolution to help improve user experience on online conference platforms such as Zoom.  
Our project aims to improve classification of satellite images by supplementing our training data with synthetic images generated via Contrastive Learning for...
A new open-source income prediction dataset along with supplemental materials that include the concept of equity in machine learning.
Exoplanet Discovery uses machine learning to help astronomers and scientists increase their knowledge of exoplanet validation algorithms.  
HealthyMamaNutrition can help millions of pregnant and lactating women link their home cooking with just-in-time recipes and nutrition guidance so they can...
Heston uses machine learning to recommend ingredient substitutions for amateur and professional chefs.  This can be used to address dietary restrictions,...
An end to end pipeline that retrieves and databases key attributes from photos of insect specimens.
Team LipSpeak is focused on helping people with voice disorders communicate effectively.
LITERATER is focused on essay grading automation to help drive improvements in literacy.
electric foresights logo
Leveraging a rich set of data sources, team Electric Foresights offers AI powered electricity price forecasting and integrated power solutions which raise the...
Our project aims to predict bat coronavirus positivity among bat species. With a collection of publicly available datasets, we fitted the models to determine...
Quotable: Making Data Come Alive!
No SQL, No problem! Use natural language to derive insights from data
w210 demo
Using multiple Machine Learning models to process NYC traffic and accident data, our team provides an accident prediction service that provides a list of...
Using object detection-based deep learning frameworks, Team SaveYourMaize is building an application that aids early and automated identification of Northern...
We performed an ablation study on the BigEarthNet dataset for the purpose of garnering a greater understanding of the relationship between feature selection...
Sustainably manage humankind’s most critical resource — freshwater.