Bill MacCartney

Overview

• Ph.D. in Computer Science from Stanford University, with focus on statistical natural language processing (NLP).
• Expertise in artificial intelligence (AI), machine learning, data mining, information retrieval, and computational linguistics.
• Extensive business experience: entrepreneurship, finance, technology development, and NLP consulting.
• Current activities: Senior Research Scientist at Google; teaching at Stanford University; occasional consulting.

Professional Experience

• Research and development of algorithms for semantic analysis of user-generated text and user modeling.

• Winter 2012: teaching CS224U: Natural Language Understanding, a grad-level course on applied computational linguistics. 50 students.
• Winter 2011: taught CS224N: Natural Language Processing, a grad-level course on statistical natural language processing. 60 students.
• Winter 2010: taught CS224U: Natural Language Understanding. 40 students.

• Research and development of back-end algorithms, with particular emphasis on machine learning models and text analysis.
• Designed algorithm used to assign topic labels to questions. Substantial contributions to question routing algorithms.
• Developed system for automatic learning of user interests from social media (Facebook and Twitter status updates).
• User modeling and preferential matching of askers & answerers with similar behavioral profiles.
• Aardvark was acquired by Google in February 2010.

• Served as consultant and advisor to various companies seeking to apply state-of-the-art NLP and machine learning techniques to everyday business problems. Worked on projects involving:
  • Extracting consumer opinions from online product reviews (sentiment analysis).
  • Developing a natural-language interface to a travel reservation system.
  • Applying machine learning methods in an online math tutoring system.

• Developed algorithms for automated question answering.
• Gained experience with web-scale technologies, e.g. MapReduce.

• Pursued research in distributed AI: strategies for query-answering over multiple (or partitioned) knowledge bases.
• Participated in developing algorithms for
  • enabling efficient collaborative reasoning among disparate KBs, and
  • automatically decomposing large KBs into minimally-connected partitions.
• Developed test suite for empirical validation and analysis of our algorithms, using a variety of real-world KBs.

• Secured $5.3M in venture funding, with Softbank as lead investor.
• Designed, built, and tested initial website in only four months, including highly-scaleable content-ranking mechanisms, browse and search caches, and an interactive animation builder.
• Website ranked in top 250 within 3 months of launch, with over 2 million unique visitors per month.
• Managed web production team. Developed efficient production process, including project queuing and prioritizing, technical standards, and project specs.
• Directly responsible for information architecture, UI design, and usability analysis. Developed multidimensional content ontology.

• Held primary responsibility for trading a $5 billion portfolio of Japanese bonds and derivatives: JGB switches, swap spreads, box trades, basis trades, listed vs. unlisted JGBs, dollar-yen basis swaps, TIBOR-LIBOR swaps, euroyen futures, Japanese municipal bonds, FX forwards and options.
• Generated and analyzed trade ideas within a relative-value (expectations arbitrage) framework.
• Utilized quantitative valuation techniques and financial modeling, including multi-factor yield curve models, no-arbitrage derivatives pricing models, and mean-reverting stochastic processes.
• Developed and maintained fixed-income analytics code in C, C++, Perl, awk, SQL, bash, Tcl.

• Led research effort to deploy Cyc AI technology in a multi-agent architecture to achieve distributed inferencing.
• Developed a C library implementing most Lisp functions to facilitate the porting of Cyc from Lisp to C.
• Ported the C version of Cyc from Unix to the Mac OS. Designed and implemented a Mac OS GUI for Cyc in C++.
• Developed HTML interface to Cyc, using cgi scripts to construct dynamic views of the Cyc knowledge base.
• Contributed to the continuing development of the Cyc knowledge base, using the CycL knowledge representation language.

• Held primary responsibility for research and curriculum development.
• Initiated program to develop and market a new LSAT course. Achieved profitability within one year.
• Developed and maintained relational database application that ran the business.
• Managed multiple course locations, with responsibility for dozens of instructors and hundreds of students.

Education

• Ph.D. in Computer Science, with focus on Natural Language Processing (NLP).
• Member of the Stanford Artificial Intelligence Lab (SAIL) and the Stanford Natural Language Processing (NLP) Group.
• Research in applying machine learning algorithms to computational semantics and natural language inference.

• Awarded Fulbright scholarship for graduate work in philosophy.
• Coursework and independent work with concentration in philosophy of language.
• Mastered German reading, writing, and speaking skills.

• Earned Bachelor of Arts in Philosophy, summa cum laude. Focus in philosophy of language and philosophy of mathematics.
• Coursework in computer science and mathematics, including mathematical logic, computability theory, differential equations.
• GPA: 4.0 in major, 3.9 overall. GRE: 800/800/800. LSAT: 180.

Awards

• Best Paper Award at the 22nd International Conference on Computational Linguistics (Coling-08), August 2008.
• Fulbright scholarship for advanced study in Germany, 1990-91.
• Dickinson Prize for best senior thesis in logic and epistemology, Princeton University, 1990.
• Phi Beta Kappa, Princeton University, 1990.

Peer-reviewed papers

An extended model of natural logic [pdf]
Bill MacCartney and Christopher D. Manning
The Eighth International Conference on Computational Semantics (IWCS-8), Tilburg, Netherlands, January 2009

A phrase-based alignment model for natural language inference [pdf]
Bill MacCartney, Michel Galley, and Christopher D. Manning
The Conference on Empirical Methods in Natural Language Processing (EMNLP-08), Honolulu, HI, October 2008

Modeling semantic containment and exclusion in natural language inference [pdf]
Bill MacCartney and Christopher D. Manning
The 22nd International Conference on Computational Linguistics (Coling-08), Manchester, UK, August 2008
—received Best Paper Award—

Natural logic for textual inference [pdf]
Bill MacCartney and Christopher D. Manning
ACL Workshop on Textual Entailment and Paraphrasing, Prague, June 2007

Learning Alignments and Leveraging Natural Logic [pdf]
Nathanael Chambers, Daniel Cer, Trond Grenager, David Hall, Chloe Kiddon, Bill MacCartney, Marie-Catherine de Marneffe, Daniel Ramage, Eric Yeh and Christopher D. Manning
ACL Workshop on Textual Entailment and Paraphrasing, Prague, June 2007

Aligning semantic graphs for textual inference and machine reading [pdf]
Marie-Catherine de Marneffe, Trond Grenager, Bill MacCartney, Daniel Cer, Daniel Ramage, Chloé Kiddon, Christopher D. Manning
AAAI Spring Symposium at Stanford, 2007

Learning to recognize features of valid textual entailments [pdf]
Bill MacCartney, Trond Grenager, Marie-Catherine de Marneffe, Daniel Cer, Christopher D. Manning
Proceedings of the Human Language Technology Conference of the North American Chapter of the Association for Computational Linguistics (HLT-NAACL 2006)

Generating Typed Dependency Parses from Phrase Structure Parses [pdf]
Marie-Catherine de Marneffe, Bill MacCartney, Christopher D. Manning
5th International Conference on Language Resources and Evaluation (LREC 2006)

Learning to distinguish valid textual entailments [pdf]
Marie-Catherine de Marneffe, Bill MacCartney, Trond Grenager, Daniel Cer, Anna Rafferty, and Christopher D. Manning
Second Pascal RTE Challenge Workshop, 2006

Robust Textual Inference using Diverse Knowledge Sources [pdf]
Rajat Raina, Aria Haghighi, Christopher Cox, Jenny Finkel, Jeff Michels, Kristina Toutanova, Bill MacCartney, Marie-Catherine de Marneffe, Christopher D. Manning, Andrew Y. Ng
Proceedings of the First PASCAL Challenges Workshop, 2005

Solving Logic Puzzles: From Robust Processing to Precise Semantics [pdf]
Iddo Lev, Bill MacCartney, Christopher D. Manning, Roger Levy
Proceedings of the ACL-04 Workshop on Text Meaning and Interpretation, July 2004

Practical Partition-Based Theorem Proving for Large Knowledge Bases [pdf, ps]
Bill MacCartney, Sheila A. McIlraith, Eyal Amir, Tomas Uribe
Proceedings of the 18th International Joint Conference on Artificial Intelligence (IJCAI-03), August 2003

The Cycic Friends Network: Getting Cyc agents to reason together [pdf, ps]
James Mayfield, Tim Finin, Rajkumar Narayanaswamy, Chetan Shah, William MacCartney & Keith Goolsbey
Proceedings of the ACM CIKM-95 Intelligent Information Agents Workshop, December 1995

Professional Service

I've served on the program committee or as a reviewer for several NLP conferences:

• Reviewer, NAACL-HLT 2007
• Reviewer, EMNLP 2008
• Reviewer, NAACL-HLT 2009
• Program Committee, TextInfer 2009
• Program Committee, NASSLLI 2010 Workshop on Inference from Text
• Program Committee, EMNLP 2010
• Program Committee, IWCS 2011
• Program Committee, TextInfer 2011
• Program Committee, EMNLP 2011

Skills

• Java, Ruby, Python, Scala, C, C++, Lisp, Perl, bash, SQL, MatLab, HTML, XML, ...
• Currently studying Cantonese. Adept (though rusty) in German and French.
• Talented writer. Widely traveled. Massively parallel. Know where Elvis is.