Bidtellect Laboratory

The Bidtellect Laboratory is an incubator to support big data analytics and digital advertising research, as well as serve as an educational platform.

Students will learn the fundamentals of real-time-bidding and be exposed to billions of programmatic transactions on a daily basis. They will learn and apply big-data technologies such as Hadoop, Kafka and Spark to build predictive models on large data sets and optimize results via real-time data streams.

About

The Bidtellect Laboratory is an incubator to support big data analytics and digital advertising research, as well as serve as an educational platform.

Students will learn the fundamentals of real-time-bidding and be exposed to billions of programmatic transactions on a daily basis. They will learn and apply big-data technologies such as Hadoop, Kafka and Spark to build predictive models on large data sets and optimize results via real-time data streams.

Bidding strategies and algorithms will be developed and tested that must support a sub 20ms execution. This will further challenge researchers and students to design new data mining and machine learning techniques.

Bidtellect has committed to not only provide guidance and access to data, but will also help deploy algorithms on their exchange. This will be particularly gratifying to students as they will have almost immediate validation of algorithm performance on a large scale, commercial platform.

About Bidtellect

---

Launched in 2013 by a group of the digital media industry’s most successful ad tech pioneers, Bidtellect is the global leader in Native Advertising technologies and solutions. The Bidtellect platform – built from the ground up to accommodate the unique challenges associated with delivery of targeted Native ads across all devices and in all formats - offers advertisers, agencies and media companies a unique and powerful toolset. Bidtellect’s proprietary state of the art technology – the most advanced in the industry today – allows native ad planning, buying, selling and overall management on a single platform. By utilizing Bidtellect’s Native DSP (nDSP), Native SSP (nSSP) and openRTB 2.3 Native Exchange, advertisers and publishers can now implement effective Native campaigns at scale with maximum optimization and ROI.

Education

Graduate Course

CAP 6807: Computational Advertising and Real-Time Data Analytics

Course Description:  This course teaches students basic concepts of computational advertising, with a focus on real-time bidding 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 bidding challenges and algorithms. The lectures will include a term project dedicated to the implementation of computational solutions to solve a real-time bidding task, using selected programming language and tools.

Course Topics:

• Computational Advertising Platforms and Marketplace
• Displaying Advertisement and Sponsored Search
• Demanding Site Platforms, Supply Side Platforms, Exchange
• Native Advertisement
• Statistical Machine Learning Algorithms: Part I: Theorems
• Statistical Machine Learning Algorithms: Part II: Applications
• Statistical Machine Learning Algorithms: Part III: Tools
• Real-time Bidding Algorithms: Click Through Rate Prediction
• Real-time Bidding Algorithms: Click Fraud Detection
• Real-time Bidding Algorithms: Bidding Curve Adjustment
• Real-time Bidding Algorithms: Customer Profiling and Retargeting
• Hands-on Term Project

Course Terms & Semesters Offered:

• 2016 Fall
• 2017 Fall

Publications

Journal

• Haicheng Tao, Xingquan Zhu, Zhiang Wu, Jie Cao, Kris Kalish and Jeremy Kayne, Online Advertising Fraud: Survey, Taxonomy, and Treatment, Wires Data Mining and Knowledge Discovery , Accepted, In Press, 2017.

Book

“Xingquan Zhu, Haicheng Tao, Zhiang Wu, Jie Cao, Kristopher Kalish, and Jeremy Kayne, Fraud Prevention in Online Digital Advertising, SpringerBriefs in Computer Science, ISBN 978-3-319-56793-8, June 2017.”