Gregory Piatetsky-Shapiro

Gregory I. Piatetsky-Shapiro (born 7 April 1958) is a data scientist, co-founder of KDD conferences and ACM SIGKDD association for Knowledge Discovery and Data Mining,[1] and President of KDnuggets, a leading site on Business Analytics, Data Mining, and Data Science. For simplicity, he usually abbreviates his name as Gregory Piatetsky.

Early life

Piatetsky was born to a Jewish-Russian family in Moscow, Russia. His father, Ilya Piatetski-Shapiro, was a well-known mathematician. He skipped 6th grade because of anti-semitic attacks and was admitted in 1970 to the Physics-Mathematics School N. 2 in Moscow.

In March 1974 Piatetsky emigrated to Israel with his mother, Inna. There, he studied mathematics at Tel-Aviv University, where at age 16 he was the youngest student. He also studied computer science for one semester at Technion, where he wrote a program in APL to play the game of Battleship. He was overwhelmingly defeated in the first game by his own program. That increased his interest in the nascent field of Machine Learning.

In 1977 Piatetsky received a full scholarship for graduate studies at NYU Courant Institute, where he received a MS (1979) and Ph.D. (1984). His first paper, published in SIGMOD in 1984, proved that secondary index selection is NP-complete by reducing it to set cover problem. However, in his dissertation he proved that the greedy method for set cover has a lower bound of 1 - 1/e ~ 63% of the optimal.

Career

After getting his Ph.D. in 1985, he joined GTE Laboratories in 1985, where he worked on intelligent interfaces to databases. In 1989 he proposed a new project at GTE called "Knowledge Discovery in Databases". The project created a number of advanced prototypes, including KEFIR (Key Findings Reporter), a system for analysis and summarization of key changes in large databases, which was a forerunner of systems like Google Analytics Intelligence. A KEFIR prototype was applied to GTE health care data and received GTE's highest technical award.

In 1997 he left GTE to join Knowledge Stream Partners (KSP), where he was Director and later Vice President and Chief Scientist. He engineered the development of component-based Knowledge Discovery Engine consulting tools, and led the consulting team that provided solutions to several banks, brokerages, telcos, insurance companies, and e-tailers.

In April 2000, KSP was acquired by Xchange, Inc, where Piatetsky's served as VP and Chief Scientist.

Piatetsky left Xchange in May 2001 and became a self-employed consultant. He works on a wide variety of projects in Business Analytics, Pharma and Life Science . He has also served as an expert witness in several cases and testified in Federal court in Manhattan on behalf of Tiffany vs eBay.

Piatetsky remains actively involved in running KDD conferences and KDnuggets.

KDD and SIGKDD

In 1989 Piatetsky organized the first workshop on Knowledge Discovery in Data (KDD-89),held at IJCAI-1989 in Detroit, MI.[1] This workshop had over 60 attendees, including researchers Ross Quinlan and Jaime Carbonell.

Piatetsky also organized the next two KDD workshops in 1991 (Anaheim) and 1993 (Washington, DC).[1] With Usama Fayyad and Ramasamy (Sam) Uthurusamy, he grew the workshops into an annual international conference on Data Mining and was the General Chair of the KDD-98 conference.

Piatetsky served as the chair of the KDD Steering committee until 1998, when the SIGKDD group was formed as part of ACM to run the annual KDD conference and help promote research in Knowledge Discovery and Data Mining. Gregory Piatetsky was elected and served as Director of SIGKDD for 2001-2005 and as SIGKDD Chair for 2005-2009.[2]

In 1997 Piatetsky and Ismail Parsa initiated the KDD Cup competition, which was the first open data mining contest in the world.

The annual ACM SIG-KDD conference is the leading research conference on Knowledge Discovery and Data Mining, according to Microsoft Academic search[3] and Google Scholar.[4] The 21st ACM SIGKDD conference was held in Sydney, Australia August 2015.

KDnuggets

KDnuggets is recognized as an important resource ("the leading newsletter"[1]) for the data mining and knowledge discovery community.

In 1993, Piatetsky started Knowledge Discovery Nuggets as a newsletter to connect researchers who attended the KDD-93 workshop. With the appearance of World Wide Web and Mosaic, Gregory and Chris Matheus created a website called Knowledge Discovery Mine, hosted at GTE Labs. The newsletter served as an unofficial publication of KDD workshops.

When Piatetsky left GTE Labs in 1997, he created KDnuggets, which stands for Knowledge Discovery Nuggets, with the mission of covering the field with short, concise "nuggets". KDnuggets started as a directory of main areas of data mining and data science, including Software, Jobs, Academic positions, CFP (calls for papers), Companies, Courses, Datasets,Education, Meetings, Publications, and Webcasts.

Now KDnuggets' main focus is to cover the news in the field of Business Analytics, Data Mining, and Data Science, including interviews with many key leaders of the field. KDnuggets also offers a free Data Mining Course designed for advanced undergraduates or first-year graduate students.

@KDnuggets Twitter was

  1. Voted the Best Big Data Tweeter by Big Data Republic (2013)
  2. Included in the 7 Business Analytics Gurus on Twitter (2013) by Information Management
  3. Named one of 10 Big Data Pros To Follow On Twitter (2014) in Information Week

In February 2015, Piatetsky and Data ScienceTech Institute announced a partnership and he became an Honorary Member of Data ScienceTech Institute Scientific Advisory Board.

Research and publications

In 1991 Piatetsky and William (Bud) Frawley edited the first book on Knowledge Discovery in Databases, and in 1996, Gregory Piatetsky, Usama Fayyad, Padhraic Smyth, and R. Uthurusamy edited a follow-up Advances in Knowledge Discovery and Data Mining

Piatetsky also helped launch and was a co-editor of Data Mining and Knowledge Discovery journal.

Piatetsky has 9 edited books and collections and over 60 technical papers, articles and book chapters, mostly focusing on data mining and knowledge discovery.

Awards and recognition

References

  1. 1 2 3 4 5 "Dr. Gregory Piatetsky-Shapiro - SIGKDD Service Award". ACM SigKDD. Retrieved 2015-09-22. Gregory Piatetsky-Shapiro has received the first ACM SIGKDD Service award for starting the KDD conferences and contributions to the KDD community, including KDnuggets newsletter. Dr. Piatetsky-Shapiro is the founder of the Knowledge Discovery in Database conference series (KDD, now the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).
  2. ↑ About SIGKDD
  3. ↑ "Top conferences in data mining". Microsoft Academic Search. Retrieved 2015-09-22.
  4. ↑ "Data Mining & Analysis". Google Scholar. Google. Retrieved 2015-09-22. 2. ACM SIGKDD International Conference on Knowledge discovery and data mining (Ranked #1 is a journal, not a conference.)
  5. ↑ Wu, Xindong (2007-09-28). "2007 IEEE ICDM Outstanding Service Award: Dr. Gregory Piatetsky-Shapiro". IEEE ICDM. Retrieved 2015-09-22. Dr. Piatetsky-Shapiro is the founder of the Knowledge Discovery in Database conference series (KDD, now the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).
This article is issued from Wikipedia - version of the Monday, February 22, 2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.