Summary of An Ontology-enabled Approach For User-centered and Knowledge-enabled Explanations Of Ai Systems, by Shruthi Chari
An Ontology-Enabled Approach For User-Centered and Knowledge-Enabled Explanations of AI Systems
by Shruthi Chari
First submitted to arxiv on: 23 Oct 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This thesis aims to bridge the gap between model-centered and user-centered explainable artificial intelligence (AI). The research focuses on creating an explanation ontology (EO) to represent different explanation types and their supporting components. A knowledge-augmented question-answering (QA) pipeline is implemented to generate contextual explanations in a clinical setting, demonstrating improved performance of large language models. Additionally, the study finds that clinicians prefer actionable information in explanations. The EO is used to represent 15 different explanation types, tested across six exemplar use cases. A system is being developed to combine explanations from various AI methods and data modalities, utilizing similarity metrics for chronic disease detection. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research helps people understand how artificial intelligence (AI) systems make decisions. It creates a special way to explain AI’s thoughts using different types of explanations. The study also makes a special tool that can help doctors understand AI’s answers in a medical setting. Doctors like this kind of explanation because it tells them what they should do next. The research shows that making AI better at explaining itself can make it more helpful. |
Keywords
» Artificial intelligence » Question answering
