This site is generated by the STRUDEL web-site management system.

I'm on sabbatical at Stanford through August 2005.

Alon Y. Halevy

Professor, Computer Science and Engineering

(was Alon Levy until 2000)

General Information

I am a professor in the Computer Science and Engineering Department at the University of Washington. I received my Ph.D in Computer Science from Stanford University , and my undergraduate degree at Hebrew University in Jerusalem in 1988. See my short bio for more detail.

I am also an entrepreneur. In 2004, I founded Transformic, a company that builds search engines for the deep web, i.e., the content on the web that is stored in databases and hides behind forms. Tranformic launched its first engine, EveryClassified.com that offers access to hundreds of classifieds sites on the web. In 1999 I co-founded Nimble Technology, one of the first companies in the space of Enterprise Information Integration. Nimble built tools for querying and integrating disparate heterogeneous data sources. Nimble was acquired by Actuate in August, 2003.

Check out the pictures of the U. of Washington Database Group Potluck Dinner

• 2001
• 2002
• 2003

Current Research

• Semex (Semantic Explorer): a tool for personal information management and integration. Semex offers users uniform access to all the personal information on their local machine (and other devices). In one seamless interface, Semex fuses information from multiple applications and from the file system. In addition, Semex is a platform for on-the-fly information integration, where users incorporate external information sources into their personal information space, thereby easily building information integration applications.
• A search engine for web services: The number of web services available is increasing rapidly, thereby posing the new search problem of finding relevant web services. We built the Woogle Web-Service Search Engine. Woogle offers novel methods for searching for web services as well as methods for detecting similarity among web-service operations.
• Peer-based data management: The Piazza Project explores the field of peer-based data management. Peer-based data management is a natural extension of data integration. Instead of having a collection of data sources and a mediator, all the peers are considered equal. They all contribute data and other resources to the system, and pose queries that require data from other peers.
• Semi-automatic schema and ontology mapping As organizations are starting sharing more and more structured data (either inernally or externally), and as the vision of the Semantic Web becomes a reality, the problem of semantically mapping between disparate data sets becomes critical. Currently, such mappings are done manually, in a labor intensive and error-prone process. We are developing a set of techniques that speed up the process of schema and ontology mapping. The techniques are based on using machine learning techniques to exploit experience from previous mappings tasks help in subsequent mapping tasks. Our LSD System was the first to embody this idea, and GLUE and Corpus-Based Matching are two subsequent systems along this line. Anhai Doan won the 2003 ACM Distingished Dissertation Award for his Ph.D thesis on this topic.
• Semantic Email: . Have you been using email for tasks that you think a database could handle for you? Semantic email attempts to solve this problem by attaching more semantics to email, where useful. See more details.

This web site is created with the Strudel web-site management system, and hence the fancy indexes on my publications below. I am fortunate to have two entries in DBLP: as a Levy and as a Halevy. These pages are often more up-to-date and more accurate than my home page.

Publications by Year

A few recent presentations

• Semex: A Platform for Personal Information Management and Integration. This particular talk was given at the NY Area DB/IR Workshop, on April 15, 2005. Earlier versions were given at Berkeley, Stanford, UC Santa Cruz, Nasa Ames and IBM Almaden in the preceding months.
• Structures, Semantics and Statistics. VLDB 2004 Keynote, Toronto, Canada, September 1, 2004.
• Crossing the Structure Chasm. Norman E. Friedmann Distingished Lecture, University of California at Los Angeles, April, 2004.
• Crossing the Structure Chasm. Distingished Lecture Series, University of British Columbia, Canada, January, 2004.
• Learning to Reconcile Semantic Heterogeneity. The New England Database Symposium (NEDS), April, 2004.
• Data Integration: Successes and Challegnes. Invited talk at IJCAI, Acapulco, Mexico, August, 2003.
• Data Integration: a Status Report. Invited talk at the German Database Conference (BTW), Leipzig, Germany, February, 2003.
• Crossing the Structure Chasm Invited talk the International Conference on Flexible Query Answering Systems, FQAS, Copenhagen, October, 2002.
• Peer-data management systems: plumbing for the semantic web. Invited talk at the AAAI Spring Symposium on Mining Answers from Text and Knowledge Bases. Stanford, California, March, 2002.
• Learning to map between ontologies. Invited talk at the IJCAI Workshop on Learning Ontologies, Seattle, Washington, August, 2001.
• Adaptive query processing. Invited talk the Canadian Database Day, Toronto, Canada, October, 2000.

Publications by Category (current through 2002)

Teaching

Check out the paper I wrote for the SIGMOD Record about the new version of my database course (published September, 2003).

• Spring 2004: CSEP 544 - Database Management Systems for the Professional Masters Program.
• Winter 2004: CSE 444 - Introduction to Database Systems.
• Winter 2003: CSE 444 - Introduction to Database Systems. This is the new and cool version, winner of the CIDR 2003 Award for Most Outrageous Idea
• Spring 2002: CSE 594 - Database Management Systems for the Professional Masters Program.
• Spring 2002: CSE 444 - Introduction to Database Systems.
• Spring, 2001: CSE 326 - Data structures and algorithms.
• Fall, 2000: CSE 594 - Database Management Systems for the Professional Masters Program.
• Spring, 2000: CSE 544 - Principles of Database Systems (the graduate database course).
• Fall, 1999: CSE 444 - Introduction to Database Systems.
• Fall, 1999: CSE 590DB - The Database Seminar: Data Mining.
• Spring, 1999: CSE 544 - graduate course on database management systems.
• Winter, 1999: CSE 473 - Introduction to Artificial Intelligence.
• Fall, 1998: CSE 444 - Introduction to Database Systems.
• Spring, 1998: CSE-590DB - graduate seminar in database systems.

My Current Graduate Students

• Luna Dong (Semex, XML Query containment, Web-service search engine)
• Jayant Madhavan (schema matching, mapping composition, corpus-based tools)
• Peter Mork (bio-informatics, knowledge representation, updates in large-scale data management systems, Piazza, Bio-Mediator)

Former Graduate Students

• Anhai Doan (University of Illinois, Urbana Champaign)
• Zack Ives (University of Pennsylvania)
• Luke McDowell (United States Naval Academy).
• Rachel Pottinger (University of British Columbia).
• Igor Tatarinov (Amazon.com).
• Marc Friedman (Microsoft SQL Server)

Funding

Other Current Professional Activities

• Program Chair of SIGMOD 2003
• Program Co-chair for the NGITS 2002.
• Editorial Board of the VLDB Journal (check out the special issue on XML query processing).
• Member of the advisory board of the Journal of Artificial Intelligence Research (JAIR).
• Associate Editor for the IEEE Bulletin on Data Engineering. Check out the special issues I edited on Adaptive Query Processing and on Query processing for XML
• Editorial Board of ACM Transactions on Internet Technology
• Guest editor for the Artificial Intelligence Journal, Special Issue on Intelligent Internet Systems (appeared in April, 2000).

Personal Information

A really good quote

Milan Kundera, The Book of Laughter and Forgetting.

Office Address:

Paul Allen Center for Computer Science and Engineering, Room 576

Mailstop 352350,

University of Washington,

Seattle, WA, 98195

Phone: (206) 543-8099

Fax: (206) 543-2969

This site is generated by the STRUDEL web-site management system.