Portrait of Fernando Diaz is unavailable

Fernando Diaz

Affiliate Member
Associate Professor, Carnegie Mellon University, School of Computer Science, Language Technologies Institutes
Adjunct Professor, McGill University, School of Computer Science
Research Scientist, Google Pittsburgh
Research Topics
Information Retrieval
Recommender Systems

Biography

Fernando Diaz is an associate professor at Carnegie Mellon University's School of Computer Science, a research scientist at Google Pittsburgh, and an adjunct professor in McGill University’s School of Computer Science.

Diaz’s expertise lies in the formal study of the search for small fragments of information in large data sets. His interests include distributed approaches to web-based documentary research, interactive and faceted research, the exploration of temporal models from news and queries, multilingual information research, and graph-based methods.

Diaz’s primary research interest is information retrieval, i.e., the formal study of searching large collections of data for small bits of information. The most familiar form of information retrieval is the web search, where users search a collection of webpages for one or a few relevant webpages. However, information retrieval goes far beyond web searches to include processes like cross-lingual retrieval, personalization, desktop search and interactive retrieval.

Diaz’s research experience includes distributed information retrieval approaches to web searching, interactive and faceted retrieval, mining of temporal patterns from news and query logs, cross-lingual information retrieval, graph-based retrieval methods, and exploiting information from multiple corpora.

For his PhD research, Diaz studied the relationship between document clustering and document scoring for retrieval using methods from machine learning and statistics. As a result, he developed an algorithm for system self-assessment and self-tuning that significantly improves the performance of retrieval algorithms across a variety of corpora.

Current Students

Haolun Wu
PhD - McGill University
Principal supervisor :
Xue (Steve) Liu
Website

Publications

On the Evaluation of Common-Sense Reasoning in Natural Language Understanding
Paul Trichelair
Ali Emami
Jackie Cheung
Adam Trischler
Kaheer Suleman
Fernando Diaz
The NLP and ML communities have long been interested in developing models capable of common-sense reasoning, and recent works have significa… (see more)ntly improved the state of the art on benchmarks like the Winograd Schema Challenge (WSC). Despite these advances, the complexity of tasks designed to test common-sense reasoning remains under-analyzed. In this paper, we make a case study of the Winograd Schema Challenge and, based on two new measures of instance-level complexity, design a protocol that both clarifies and qualifies the results of previous work. Our protocol accounts for the WSC's limited size and variable instance difficulty, properties common to other common-sense benchmarks. Accounting for these properties when assessing model results may prevent unjustified conclusions.
2018-11-05
arXiv.org (preprint)
dblp.uni-trier.de
Advances in Information Retrieval
Diane Kelly
Fernando Diaz
Nicholas J. Belkin
James Allan
2004-04-05
Lecture Notes in Computer Science (published)
doi.org
Advances in Information Retrieval
Diane Kelly
Fernando Diaz
Nicholas J. Belkin
James Allan
2004-01-01
ECIR (published)
doi.org