Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn how to evaluate advanced search features in professional domains through a comprehensive conference talk examining AMBOSS's implementation of knowledge panels, preview snippets, and vector search capabilities. Discover the unique challenges of measuring search effectiveness in specialized environments where traditional e-commerce metrics like purchases or basket additions don't apply, particularly in medical information platforms serving 60% of US medical students and hundreds of thousands of doctors worldwide. Explore alternative evaluation frameworks that combine user feedback loops and engagement-based heuristics to assess search feature performance through A/B testing methodologies. Understand why click-through rates can be misleading indicators depending on context, serving as signals for both relevant results and irrelevant answer snippets. Gain practical insights into developing robust measurement strategies for algorithmic improvements and user experience changes in high-stakes, knowledge-intensive search environments where accuracy and relevance are critical for professional decision-making.
Syllabus
Haystack US 2025 - Valentin v. Seggern: Evaluating Search Features
Taught by
OpenSource Connections