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.
CHASERS OF THE LOST DATA | Everything's not lost
Improve information processing by imputing the lost data that will later be used in ML models in order to enhance the accuracy of predictions or classification models.
BACKGROUND
- NASA has sensors that take information from space and through Telemetry is transmitted to a central station that analyze the data.
- Due to external events such as cosmic rays, electromagnetic pulses and others, cause loss of connectivity between the NASA sensors and the central stations, some data does not arrive completely causing the loss of relevant information.
WHAT WE DID?
- Improve information processing by imputing the lost data that will later be used in ML models in order to improve the accuracy of predictions or classification models.
- We focus on the design of a architecture capable of receiving data from NASA's electronic instruments, and impute the missing data for maximize the most important features that the ML engineer needs to be able to develop projects with great performance.
- The system is not only focuses on improving the loss of data from NASA's instruments but also engages in the improvement of any electronic device capable of acquiring data from its environment.