Abstract
Architecture framework reviews (AFRs) are valuable tools offered by cloud infrastructure providers to enable customers to validate their implementation against a set of curated architectural best practices. AFRs are currently offered via self-service questionnaires, which tend to be inflexible, or via reviews by human experts, which, although guided, are less accessible and more expensive. This disclosure describes a conversational artificial intelligence (AI) interface (chatbot) that enables dialog-based architecture framework reviews and alignment assessment. The described automated self-service tool has natural language capabilities that enable dialog-based interactions and guidance, and draws from a pool of static questions to pose architectural framework review questions (AFRQ). The user provides answers in a natural language, unstructured format. The answers are interpreted by AI using natural language understanding (NLU) and are mapped to a level of alignment or maturity. Since NLU is used, an exact text match or static logical mapping are not required.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Dantas, Victor, "A Conversational AI Approach to Architecture Framework Reviews", Technical Disclosure Commons, (February 08, 2023)
https://www.tdcommons.org/dpubs_series/5672