Quality Assurance of Generative Dialog Models in an Evolving Conversational Agent Used for Swedish Language Practice
(2022) 1st International Conference on AI Engineering - Software Engineering for AI, CAIN 2022 In Proceedings - 1st International Conference on AI Engineering - Software Engineering for AI, CAIN 2022 p.22-32- Abstract
Due to the migration megatrend, efficient and effective second-language acquisition is vital. One proposed solution involves AI-enabled conversational agents for person-centered interactive language practice. We present results from ongoing action research targeting quality assurance of proprietary generative dialog models trained for virtual job interviews. The action team elicited a set of 38 requirements for which we designed corresponding automated test cases for 15 of particular interest to the evolving solution. Our results show that six of the test case designs can detect meaningful differences between candidate models. While quality assurance of natural language processing applications is complex, we provide initial steps toward... (More)
Due to the migration megatrend, efficient and effective second-language acquisition is vital. One proposed solution involves AI-enabled conversational agents for person-centered interactive language practice. We present results from ongoing action research targeting quality assurance of proprietary generative dialog models trained for virtual job interviews. The action team elicited a set of 38 requirements for which we designed corresponding automated test cases for 15 of particular interest to the evolving solution. Our results show that six of the test case designs can detect meaningful differences between candidate models. While quality assurance of natural language processing applications is complex, we provide initial steps toward an automated framework for machine learning model selection in the context of an evolving conversational agent. Future work will focus on model selection in an MLOps setting.
(Less)
- author
- Borg, Markus LU ; Bengtsson, Johan ; Osterling, Harald ; Hagelborn, Alexander ; Gagner, Isabella and Tomaszewski, Piotr
- publishing date
- 2022
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- action research, AI quality, conversational agent, generative dialog model, requirements engineering, software testing
- host publication
- Proceedings - 1st International Conference on AI Engineering - Software Engineering for AI, CAIN 2022
- series title
- Proceedings - 1st International Conference on AI Engineering - Software Engineering for AI, CAIN 2022
- pages
- 11 pages
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- 1st International Conference on AI Engineering - Software Engineering for AI, CAIN 2022
- conference location
- Pittsburgh, United States
- conference dates
- 2022-05-16 - 2022-05-17
- external identifiers
-
- scopus:85133467455
- ISBN
- 9781450392754
- DOI
- 10.1145/3522664.3528592
- language
- English
- LU publication?
- no
- id
- 9ebe54d7-4156-4ba8-bc0c-ed6af761f908
- date added to LUP
- 2022-10-24 14:57:11
- date last changed
- 2022-10-24 14:57:11
@inproceedings{9ebe54d7-4156-4ba8-bc0c-ed6af761f908, abstract = {{<p>Due to the migration megatrend, efficient and effective second-language acquisition is vital. One proposed solution involves AI-enabled conversational agents for person-centered interactive language practice. We present results from ongoing action research targeting quality assurance of proprietary generative dialog models trained for virtual job interviews. The action team elicited a set of 38 requirements for which we designed corresponding automated test cases for 15 of particular interest to the evolving solution. Our results show that six of the test case designs can detect meaningful differences between candidate models. While quality assurance of natural language processing applications is complex, we provide initial steps toward an automated framework for machine learning model selection in the context of an evolving conversational agent. Future work will focus on model selection in an MLOps setting.</p>}}, author = {{Borg, Markus and Bengtsson, Johan and Osterling, Harald and Hagelborn, Alexander and Gagner, Isabella and Tomaszewski, Piotr}}, booktitle = {{Proceedings - 1st International Conference on AI Engineering - Software Engineering for AI, CAIN 2022}}, isbn = {{9781450392754}}, keywords = {{action research; AI quality; conversational agent; generative dialog model; requirements engineering; software testing}}, language = {{eng}}, pages = {{22--32}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, series = {{Proceedings - 1st International Conference on AI Engineering - Software Engineering for AI, CAIN 2022}}, title = {{Quality Assurance of Generative Dialog Models in an Evolving Conversational Agent Used for Swedish Language Practice}}, url = {{http://dx.doi.org/10.1145/3522664.3528592}}, doi = {{10.1145/3522664.3528592}}, year = {{2022}}, }