Data Mining and Electronic Business

Andreas S. Weigend, Stanford University

Course: Statistics 252. 3 units. Summer 2003.

Description: The Internet and related technologies have caused the cost of communication and transactions to plummet, and consequently the size of potentially relevant data sets to explode. This course discusses some of the underlying principles and statistical issues, presents algorithmic approaches to data mining and e-business, and describes some real-world examples.

Course Topics:
- Session analysis
- Shopping process models
- Identity and reputation management
- Recommender systems
- Search engines
- Design of Web experiments
- Pricing and promotions

Time: MTW 3:15pm - 5:00pm. We are fortunate to have sessions have several guest speaker joining us us (see schedule).

Dates: Tuesday June 24 (first class) through Wednesday July 17 (last class). Final assignment due Saturday July 19.

Location: Skilling 193. This course is also available over the Web through SPCD, scpd.stanford.edu/scpd/courses/academic/sched_summer2003.htm

TA: Saharon ROSSET (saharon@stanford.edu). Please use email to contact the TA with questions and to set up appointments.

Textbook: Baldi, Frasconi and Smyth: Modeling the Internet. Wiley, 2003.

Additional Readings: Will be available on the Web through links from the syllabus. (url: www.weigend.com/Teaching/Stanford/Readings).

Slildes: Will be available on the Web through links from the syllabus. (url: www.weigend.com/Teaching/Stanford/Slides).

Draft Schedule:

		Guest	Read BEFORE class
(1) T 6/24	Overview / Expectations
(2) W 6/25	Pricing	Paat Rusmevichientong
R 6/26	<Optional: Background>	Location: Library, Statistics Department	Ch 1-2
(3) M 6/30	Search	Jan Pedersen	Ch 3-5	HW1 (KDD Cup) out
(4) T 7/1	Insights from Blue Martini	Ronny Kohavi	Kohavi et al (KDD cup paper)
(5) W 7/2	Session clustering		Ch 7, Webcanvas, MIT HMM , CMU or UCLA paper
(6) M 7/7	Clickstream analysis	Bruce D’Ambrosio	Bruce RPM draft, Getoor bckgrnd	HW1 due
(7) T 7/8	Customer model HW1 discussion
(8) W 7/9	Customer behavior (B2C)
(9) M 7/14	Experimental design	Art Owen		HW2 out
(10) T 7/15	Recommender systems Electronic marketplaces		Ch 8 Shani, Brafman and Heckerman (2002) Heckerman et al (2000)
(11) W 7/16	Social networks Outlook		Domingos and Richardson (2002)	HW2 due Sat 7/19

Faculty: Andreas S. Weigend is the Chief Scientist at Amazon.com where he is responsible for research in machine learning and computational marketing. Applications range from real-time predictions of customer intent and satisfaction, to long-term optimization of pricing and promotions.
Besides this course at Stanford, he teaches at CEIBS (China Europe International Business School. Shanghai) in the executive program. Previously, he was full-time faculty at New York University's Stern School of Business and at the University of Colorado at Boulder. He has published more than one hundred scientific papers and co-authored six books.
His entrepreneurial career includes Moodlogic, recently voted the best music organizer by cnet, and ShockMarket, a financial market data analysis firm backed by D.E. Shaw and Deutsche Bank. He has consulted to many information-rich organizations, including Acxiom, Bank of America, Bertelsmann Venture Capital, Goldman Sachs, J.P. Morgan, Morgan Stanley, Siemens, Union Bank of Switzerland, Vividence, and others.
He received his Ph.D. from Stanford University in physics in 1991, and was a researcher at Xerox PARC (Palo Alto Research Center) and the Santa Fe Institute. He studied electrical engineering, physics, and philosophy at Karlsruhe University, Bonn University (Germany), and Trinity College, Cambridge (U.K.).

by aweigend at stanford.edu