CAP 6807: Computational Advertising and Real-Time Data Analytics
Announcement
(The content of this page changes frequently)
Class Time: Tuesday: 4:20PM - 7:00PM
Classroom: CM 128
Office Hour: Tuesday: 1:00PM - 3:00PM
Textbook:
Reference Materials:
- Programmatic Advertising: The successful transformation to automated, data-driven marketing in real-time , Springer, 2016
- Machine Learning in Online Advertising , NIPS 2010 Workshop
- Internet Advertising and the Generalized Second-Price Auction:Selling Billions of Dollars Worth of Keywords, American Economic Review, American Economic Association, 97(1):242-259, 2007
- Computational Advertising: Techniques for Targeting Relevant Ads , Foundations and Trends in Information Retrieval: 8(4-5):263-418, 2014
Other Useful Resources:
Course Description:
On-line digital advertising is quickly dominating the advertising market, and its market size is projected to reach over $250 billion in 2018. This course teaches students basic concepts of computational advertising, with a focus on real-time big data analytics for displaying advertisement. The class will introduce different key aspects of building platforms for online advertising, the computational requirement, tools, and solutions. The class will cover three major topics including (1) basic statistical machine learning and data analytics skills, (2) Display advertising platforms, tools, and domain knowledge; and (3) Real-time big data analytics (including Hadoop and Spark). The lectures will include a term project dedicated to the implementation of computational solutions to solve an analytics task, using selected programming language and tools.
Topics:
- Display advertising platforms, tools, and domain knowledge
- Introduction to Advertising
- Computational advertising platforms and marketplace
- Displaying advertisement, sponsored search
- Demanding site platforms, supply side platforms, Exchange
- Statistical machine learning and data analytics skills
- Intro to statistical machine learning algorithms
- Logistic regression and Classification
- Statistic machine learning tools: R
- Real-time big data analytics algorithms
- MapReduce Programming
- Spark Programming
- Real-time bidding algorithms: Click through rate prediction (Spark)
- Real-time bidding algorithms: Click fraud detection (Spark)
Lectures, Assignments, and Projects
Course Schedule by Week
Resources
Homeworks
Dataset
Project
Grading policy:
Homework |
35 |
Mid-term |
15 |
Course project |
35 |
Student Presentation |
10 |
Participation |
5 |
Your final grade will be based on the scores you have earned from the above categories (compared to the performance of other students in the class).
Project: The goal of the term project is to practice analytical skills learned from the class to solve real-world computational advertising and real-time data analytics challenges.
The instructor will help each student identify a suitable topic (a set of tentative topics, such as click through rate prediction, will be distributed in the class). Students are required to apply knowledge learned from the class to solve the identify task, implement and validate the design, and collect experimental results for reporting.
The final outcomes of the project will be turned into a 6-8 page double column technical report.
Late policy:
All assignments are due midnight on the assigned due date. Please refer to the Assignments and Projects for details on submission. Late submission is allowable, however, the late penalty is -2pts/day.
Communication:
All important course communication will be done using your fau.edu email address. Sending email to me from another account is disencouraged, and if you do you must set the reply-to field to your FAU email account if the message concerns grading or evaluation in any way. You must also include your name in all messages concerning the course.
If you have your FAU email forwarded to an AOL or other email account, read this important notice concerning blocking of FAU email.
All work in this course must be INVIDUAL effort unless specified otherwise.