CAS CS 591 - Fall 2015 - Electronic Commerce
Surprisingly, until very recently, the world of electronic commerce has received
relatively little attention in academia. It is true
that economists and management scientists have studied the characteristics
of successful and unsuccessful Internet firms. Moreover, game theorists
and computer scientists have helped pioneer foundations such as the theory of
second-price auctions. Finally, innovation in distributed systems that power
access to massive amounts of data have facilitated the realization of
the massive two-sided markets driven by search engine advertising and ad networks.
However, only recently
have experimentalists deeply tapped into the opportunity that unlike traditional markets,
many aspects of the electronic commerce marketplace are not only publicly
observable, but are readily available for online measurement and data collection.
Therefore, research on questions such as the prevalence of "sniping" on eBay,
the effectiveness of Groupon personalizing its daily deals for subscribers, or
the study of how landlords learn how to price inventory on AirBnB, can all be
evaluated via large-scale measurements, enabling studies that were not previously
In this class, we will consider electronic commerce from a broad and
inter-disciplinary perspective, reading seminal papers on theoretical
foundations and empirical findings written by the Computer Science, Information
Systems, Marketing, and Economics
communities; we plan to attract course participants and guest lecturers from
these various disciplines. Our goal, however, will then be to focus on quantitative
evaluation of the e-commerce marketplace, and to enable students to conduct
research in this area. A core competency that we will develop is fluency with big
data: experimental methods; best practices and techniques for data collection, data
mining, and statistical analysis; effective presentation of findings; as well as the
ethics of data collection. The capstone project of the course will be a research
project, conducted by individuals or in pairs, in which students conduct a quantitative
measurement-driven analysis of a computational aspect of an e-commerce firm or of consumer
behavior with respect to an e-commerce marketplace.
Prerequisites: This course is designed for students who are potentially interested in pursuing graduate research related to Electronic Commerce. Please note that this course is not about entrepreneurship. While students' backgrounds will vary, it is expected that students have completed an undergraduate major in an area related to the course topics (such as CS, Economics, or Marketing). Seniors who have completed all required coursework except for electives should seek the instructor's permission to enroll.
Prof. John W. Byers
Email: byers @ cs . bu . edu [preferred]
Phone: 617-353-8925 [do not leave voice-mail; use e-mail instead]
Office Hours (Fall 2015):
Open hours: Tues 2-4
By prior appointment only: Wed 10-11.
Academic: John is Professor of Computer Science at Boston University. His academic research interests are broadly focused on algorithmic and economic aspects of e-commerce, networking, and large-scale data management. His work strikes a balance between
theoretical foundations and rigorous data-driven experimentation.
Entrepreneurial: John is founding Chief Scientist and Board Member at Cogo Labs, a start-up based in Cambridge, MA, where
he has had an executive role since the company's founding in 2005. Cogo leverages a unique proprietary
technology platform for algorithmic marketing, data mining, and quantitative business analytics to guide incubated
portfolio companies from inception to profitability and beyond.
Class meeting time: Tues/Thurs 12:30-2:00, MCS 180 (Hariri Institute Conference Room).
*We will use our registrar-assigned classroom, MCS B23, only on day one and when the Hariri room is in use.*
Course Requirements and Grading:
There will be three components of the grade in the class:
For class, we will be drawing on some material from the Easley-Kleinberg textbook (see below), but more often, we will be reading and discussing research papers. I will also be giving some lectures on technical background material for methods used in the papers. In the paper-reading portion of the course, students will be required to read and digest
approximately two papers per week, prior to lecture.
Students will submit short summaries and provide answers to basic questions about the papers prior to discussion.
For each major topic of the course, a group of students chosen in advance will serve as specialists on the topic --
they will be experts on the papers we are discussing, and will be expected to help facilitate the discussion,
brainstorm about research directions, and help with the presentation of the material (or with supplemental material).
- Summarization and in-class discussion of the readings in the course (25%).
- Quizzes and homework assignments (25%).
- Comprehensive, semester-long research project (50%).
We will have periodic short assignments, two in-class quizzes comprising short answer problems, and perhaps a few
longer homework problems.
The capstone project for the course will be a semester-long research project, culminating in a writeup in the
style of a conference paper, and a presentation to the class, which most likely will take the form of a poster
at a class-wide poster session. The topic of the research project will be for students to conduct a
quantitative measurement-driven analysis of a computational aspect of an e-commerce firm or of consumer behavior
with respect to an e-commerce marketplace. Students may work alone or in teams of two, with the expected output
of the teams to be commensurately larger. Suggested project topics and project deadlines will be announced
after the first few weeks of the course. I will expect students in this class to take the project very
seriously and there will be regular interaction with the instructor outside of class to work on the projects ---
ideally, several of the projects in the class will eventually lead to publishable papers. A strong venue
for Computer Science students to target could be the experimental track of
the ACM Symposium on Economics and Computation.
For economics students, the goal of the project would be to write a paper that could develop into a chapter of
the dissertation and potentially a job market paper. Ideally, the ideas in the paper could be developed into
work publishable at a top field or general interest journal.
- Course overview. Technical backdrop; e-commerce challenges and opportunities.
- Data analytics and useful statistical methods.
- Methods and practice of large-scale data collection.
- Auction design and mechanism design.
- User-generated content: mining, valuing, securing.
- Reputation and branding.
- Recommender systems and personalization.
- Social networks and user data.
- Ad auctions and search engine advertising.
- Emerging e-commerce business models: cookie tracking, targeted advertising.
Course Topics, Handouts, and Readings
- [Thurs 9/3]: Course syllabus and overview of E-commerce.
- [Tues 9/8, Thurs 9/10]: A/B testing and running controlled experiments.
"Controlled experiments on the web: survey and practical guide",
R. Kohavi, R. Longbotham, D. Sommerfield, R. Henne, DMKD (18) 140-181, 2009.
Assignment 2 and associated data files
Here's a link to one of the
student-submitted Sage worksheets.
- [Tues 9/15, Thurs 9/17]: Data analytics with Sage and R on Sage Math Cloud.
Two example Sage worksheets (ex1,
ex2) were presented.
Background material on Auctions, Chapter 9 from
"Networks, Crowds, and Markets: Reasoning About a Highly Connected World",
David Easley and Jon Kleinberg. Cambridge University Press, 2010.
version is available on the authors' website, or purchase a copy.
- [Tues 9/22]: Dan Ariely, Axel Ockenfels, and Alvin E. Roth,
Analysis of Ending Rules in Internet Auctions", Rand Journal of Economics, 36, 4, Winter 2005, 891-908.
[Thurs 9/24]: John W. Byers, Michael Mitzenmacher and Georgios Zervas,
"Information Asymmetries in Pay-Per-Bid Auctions",
Proc. of 11th ACM Conf. on Electronic Commerce (EC '10),
Cambridge MA, June 2010.
[Tues 9/29]: Background on linear regression methods. Theoretical basis for ordinary least squares (OLS) and maximum likelihood estimation (MLE) approaches. Handling binary, categorical and ordinal variables. Quick mention of logistic and probit models, which we'll see again later.
Michael Luca, "Reviews, Reputation, and Revenue:
The Case of Yelp.com." Harvard Business School Working Paper, No. 12-016, September 2011.
[10/6]: Prof. Zervas from Questrom @ BU was on hand to discuss his paper:
Michael Luca and Georgios Zervas, "Fake
It Till You Make It: Reputation, Competition, and Yelp Review Fraud," forthcoming,
[10/8]: Discussion of Assignment 3, and the related
Rating Analysis on Review Text Data: A Rating Regression Approach," H. Wang, Y. Lu and C.X. Zhai,
17th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2010).
Handout: Short description of our
[10/13]: No class, BU is on a Monday schedule.
[10/15]: Continued discussion of last week's LARA paper.
Then begin discussion of: The Groupon Effect on
Yelp Ratings: A Root Cause Analysis,
John W. Byers, Michael Mitzenmacher and Georgios Zervas.
In Proc. of the 13th ACM Conference
on Electronic Commerce (EC '12), Valencia, Spain, June 2012.
Presentation slides (in Keynote).
[10/20]: Continued discussion of the Groupon paper.
Then: Matching markets, and connections to auctions. Market-clearing prices and properties of market-clearing allocations. A good basic reference is Chapter 10
of the Easley-Kleinberg text.
[10/22]: In-class quiz (50 minutes), then more on matching markets.
[10/27]: Quiz answers, wrap up of matching markets. Then: connection to
sponsored search (i.e., search engine advertising). Chapter 15 in Easley-Kleinberg.
[10/29]: Short project writeups due, class discussion.
Then we continued dicussion of Chapter 15 in Easley-Kleinberg.
[11/2]: Wrapup of Chapter 15: Proof that VCG pricing facilitates truthful revelation. GSP vs. VCG.
Then, empirical work on search advertising:
"An Empirical Analysis of Search Engine
Advertising: Sponsored Search in Electronic Markets", by Ghose and Yang, Management Science, 55(10),
2009, pp. 1605-22.
[11/4]: Ghose/Yang paper, then click fraud and how to measure it.
and Fingerprinting Click-Spam in Ad Networks", by Dave, Guha, and Zhang,
Proc. of ACM SIGCOMM 2012.
- [11/10]: Recommender systems: overview of objectives and basic methods.
The 2005 survey
by Adomavicius and Tuzhilin, IEEE TKDE 17(6), June 2005, still serves as a useful
- [11/12] Methods used by the BellKor team to win the Netflix
Prize, starting from their KDD '07 paper:
Relationships at Multiple Scales to Improve Accuracy of Large Recommender Systems",
by R. Bell, Y. Koren, and C. Volinsky.
Several good slide decks about the Netflix prize are out there, including from
- [11/17] John in Korea. Class cancelled.
- [11/19] Davide Proserpio will be on hand to present his recent work on Airbnb:
"The Rise of the Sharing Economy:
Estimating the Impact of Airbnb on the Hotel Industry", by G. Zervas, D. Proserpio and J. W. Byers, ACM EC '15.
- [11/24] Second-round versions of class project writeups due. Davide and I presented some material on how to make and present a good project poster. We then gave each team a few minutes of air-time to present a status update.
- [11/26] Happy Thanksgiving!
- [12/1] We concluded our discussion of the Netflix Prize and had team project meetings.
- [12/3] We covered
Visible Hand: Race and Online Market Outcomes", by J. Doleac and L. Stein.
- [12/10] We'll finish up by reflecting on this short overview paper in light of what we've studied:
J. Feigenbaum, D. Parkes, and D. Pennock,
in E-Commerce", CACM, January 2009, 52(1), pp. 70-74.
One topic we didn't get to was prediction markets. We might dip into this paper a little bit if time permits.
Using Prediction Markets to Track Information Flows,
by Cowgill, Wolfers, and Zitzewitz.
- [Fri, 12/11] Poster session, 1-4PM, Hariri.
We have compiled a bunch of poster examples in this
directory for you to peruse.
Please follow the directions to print your poster on the
CS Wiki page.
You will need to submit your poster by 9PM Tuesday to ensure that it's ready for Friday.