Accession Number:
AD1147964
Title:
MODSQUAD, The Purdue Data-Driven Discovery of Models Team
Descriptive Note:
[Technical Report, Final Report]
Corporate Author:
Purdue University
Personal Author(s):
Report Date:
2021-09-17
Pagination or Media Count:
57
Abstract:
Our team consisted of both Statistics and Software Design experts. Our involvement in the DARPA Data-Driven Discovery of Models D3M program included developing a TA3 system, which we entitled ModSquad, participating in joint standards design meetings, presenting at the project workshops on a variety of topics, and investigating the TERRA-REF datasets and challenge problems, resulting in an interactive data exploration and model fitting interactive interface for several TERRA-REF datasets, and finally culminating with an analysis of state-of-the-art atmospheric research data. We feel that we accomplished the tasks assigned to our team to the benefit of the overall program and provided a high return on investment for DARPA, AFRL, and the D3M program leadership.
Descriptors:
- information science
- air force
- data science
- machine learning
- neural networks
- air force research laboratories
- computer languages
- computer programming
- data analysis
- detection
- engineering
- governments
- united states
- water vapor
- domain specific programming languages
- software development
- user interface
- data mining
- databases
- environment
Subject Categories:
- Computer Programming and Software
- Cybernetics
- Information Science