Workshop on Big Data & Data Science in Retail


ICDM 2017: IEEE International

Conference on DATA MINING 2017


New Orleans, Louisiana, USA

November 18-21, 2017

Important Dates:

All Deadlines are at 11:59PM Pacific Daylight Time.

Note: Workshop Paper Submission Extension to August 14th

Workshop Paper


Workshop Paper


Camera-Ready Final

of Accepted Papers



Big Data and Data Science in Retail

Half Day Workshop

The rapidly changing landscape of technology is creating new opportunities and challenges for retailers.  New data sources coupled with traditional retail data unleash the potential for innovative solutions in the retail industry.

Broadly speaking, retailers consider problems across two key domains: 1) Merchandising and Operations and 2) Marketing. Whereas the former focuses largely on product assortments, pricing and mass promotional decisions, and inventory and supply chain management, the latter focuses on promoting awareness and improving overall customer experience. Data mining and statistics-driven decision making have been the keys to success in both these domains.

However, retail data has increased exponentially in volume, variety, and velocity with every passing year. This includes both traditional retail data (e.g. transactional sales, inventory and logistics, and customer loyalty, etc.), as well as “newer” data sources from online, mobile (e.g. apps, IoT, etc.) and other external sources such as social and real-time data (e.g. weather, satellite imaging etc.).

Coupled with advancements in data and computing systems, the application of big data tools and machine learning techniques to this plethora of retail data offers exciting new opportunities to develop competitive solutions for innovative retailers.

Accepted Papers

Mobile E-Commerce Data Processing Using Relational Memory

Authors: Parham Aarabi

Improving Multivariate Time Series Forecasting with Random Walks with Restarts on Causality Graphs

Authors: Piotr Przymus, Youssef Hmamouche, Alain Casali, and Lotfi Lakhal

A Pattern Tree based Method for Mining Conditional Contrast Pattern from Multi-Source Data

Authors: Li Li, Sarah Erfani, and Christopher Andrew Leckie

Keynote Speakers

Philippe Beaudoin, PhD

Element AI (SVP Research Group)

Philippe cofounded Element AI in 2016 and currently leads its applied and fundamental research groups. His team has helped tackle some of the biggest and most interesting business challenges using machine learning. Philippe's prior explorations included multidimensional time-series analysis during his PhD at Université de Montréal, and bipedal walking control as a postdoc at UBC. He also worked five years at Google as a Senior Developer and Technical Lead Manager, partly with the Chrome Machine Learning team. When he has some free time, Philippe likes to invent new boardgames, and sometimes get them published ;).

Brian Keng, PhD

Rubikloud Technologies

Brian is the Chief Data Scientist at Rubikloud Technologies where he leads a team building out intelligent enterprise solutions for some of the world's largest retail organizations. Brian is a big fan of Bayesian statistics but his main professional focus is around building out scalable machine learning systems that seamlessly integrate into traditional software solutions. Before Rubikloud, Brian has worked at Sysomos leading a team of data scientists performing large scale social media analytics working with datasets such as the Twitter firehouse. He earned his PhD in Computer Engineering from the University of Toronto during which time he was an early employee of a startup that commercialized some of his research.


Ed Kim

VP Insights, Indigo

Waleed Ayoub

Chief Product Officer, Rubikloud

Brian Keng

Chief Data Scientist, Rubikloud

Kanchana Padmanabhan

Senior Data Scientist, Rubikloud 

Program Committee

Dr. Ayse Bener

Ryerson University

Dr. Zhengzhang Chen

NEC Laboratories America, Inc

Dr. Andreas Veneris

University of Toronto

Dr. Amir Nabatchian


Dr. Andriy Miranskyy

Ryerson University

Dr. Arthur Ryman

Ryerson University

Dr. Tamer Abdou

Ryerson University

Dr. Uzair Ahmad

Ryerson University

Dr. Ceni Babaoglu

Ryerson University

Dr. Pablo Hennings


Dr. Sebnem Kuzulugil

Ryerson University

Dr. Koushik Pal


Dr. Robert Chin

Freckle IoT, Canada


8:30 AM


8:40 - 9:05 AM

Keynote by Brian Keng PhD, Chief Data Scientist, Rubikloud

9;10 - 9:30 AM

(SP20206) Improving Multivariate Time Series Forecasting with Random Walks with Restarts on Causality Graphs

9:35 - 9:55 AM

(SP20202) "Mobile E-Commerce Data Processing Using Relational Memory"

10:00 - 10:15 AM


10:15 - 11:15 AM

Keynote by Philippe Beaudoin, PhD, Element AI (SVP Research Group)

11:20 - 11:40 AM

(DM513) A Pattern Tree based Method for Mining Conditional Contrast Pattern from Multi-Source Data

The workshop is free for all who register to attend ICDM 2017


The Roosevelt New Orleans

130 Roosevelt Way,

New Orleans,

LA 70112, USA