Fernando Diaz

Fernando Diaz's research focuses on the design and evaluation of information access systems, including information retrieval (e.g. web search) and recommender systems. This has included developing evaluation metrics, designing algorithms, and understanding the broader, societal implications of these technologies.

My recent projects include:

• Evaluation
  • preference-based evaluation: improving the sensitivity of evaluation metrics by measuring pairwise preferences between systems
  • fairness in ranking: designing evaluation metrics for ranked lists of items capturing different notions of fairness
  • artificial intelligence and culture: understanding and measuring the impact of artificial intelligence on culture industries such as music, film, and literature
• Algorithm Design
  • retrieval-enhanced machine learning: designing algorithms to support machine learning systems
  • tip of the tongue retrieval: designing algorithms to help people re-find previously consumed content

His previous projects have included text summarization in crisis informatics contexts, inferring user intent from mouse cursor behavior, aggregated search, time-sensitive ranking, and core retrieval algorithms.