Can AI Close the Design Code Abstraction Gap?

Abstract

We are motivated to help create a new generation of automation for architects that helps bridge the gap between architecture abstractions and code. We are encouraged by potential applications of AI to detect design abstractions (allows us to recover "as implemented" designs and check that implementation conform to "as intended" designs) and refactor code to improve its design.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 26, 2019
Accession Number
AD1085210

Entities

People

  • Ipek Ozkaya
  • James Ivers
  • Robert Nord

Organizations

  • Carnegie Mellon University

Tags

DTIC Thesaurus Topics

  • Automation
  • Commerce
  • Contractors
  • Contracts
  • Demographic Cohorts
  • Department Of Defense
  • Engineering
  • Governments
  • Guarantees
  • Materials
  • Platforms
  • Software Development
  • Universities

Fields of Study

  • Computer science
  • Engineering

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Geospatial Intelligence and Artificial Intelligence Analytics
  • Systems Analysis and Design