Knowledge Panels

There's currently very little academic research on knowledge panels, especially on how users interact with them. There is, however, plenty of work being done on search interfaces generally, including on very specific topics such as search interfaces for children and search interfaces for medical records. More relevant for LD4P2 is the research being done on natural language with semantic searches:

  1. SWSNL: Semantic Web Search Using Natural Language
  2. Natural language data management and interfaces
  3. Searching and browsing Linked Data with SWSE: The Semantic Web Search Engine

Research on knowledge graphs can provide some insights into search interface usability issues:

  1. Evaluating search features of Google Knowledge Graph and Bing Satori: Entity types, list searches and query interfaces
  2. What are the Important Properties of an Entity? Comparing Users and Knowledge Graph Point of View

In this second article, the authors reverse-engineered the Google Knowledge Pa nel to identify what Google thinks are the most important properties to include on a panel. They then compared these results to a user survey, which documented user expectations, and found that Google and these users agreed about 55 percent of the time.

  • No labels

All content on the LYRASIS Wiki is licensed under the CC BY (Attribution) license, unless otherwise noted.