Handlny| Chasers of the Lost Data

Project Details

The Challenge | Chasers of the Lost Data

Help find ways to improve the performance of machine learning and predictive models by filling in gaps in the datasets prior to model training. This entails finding methods to computationally recover or approximate data that is missing due to sensor issues or signal noise that compromises experimental data collection. This work is inspired by data collection during additive manufacturing (AM) processes where sensors capture build characteristics in-situ, but it has applications across many NASA domains.

Handlny

Video could not be embeded.Watch Here

Data in the real-world are rarely clean. Our idea is to build a library that infers the missing data through 2 approaches. The first one is by using realistic methods such as using equation. The second one is through deep learning.

Handlny

Background

Data in the real-world are rarely clean. Our idea is to build a library that infers the missing data through 2 approaches. The first one is by using realistic methods such as using equation. The second one is through deep learning.


Resources

We tested our techniques on Meteor Landing and Fireballs datasets. Those datasets have missing data in latitude, longitude, speed and different other fields.

Meteor Landing

https://catalog.data.gov/dataset/meteorite-landing

Fireballs and bolide

https://catalog.data.gov/dataset/fireball-and-bolide-reports


Challenges

Our biggest challenge was time. Understanding the Imputation techniques and their complexities was also another challenge as most techniques needs a very strong mathematical background.

Github Link

https://github.com/Nada-ibrahim/Handlny-Imputation-Package