Tutorial 5: Introduction to Business Analytics

Presented by

Kush R. Varshney


This tutorial provides an introduction to business analytics for information engineers with no prior business administration experience. Business analytics is the application of statistical and signal processing methods to gain insight about business performance and drive business planning.

The world has changed with globalization, instantaneous communication, expansive enterprises, and an explosion of data and signals along with ample computation to process them. In the present age, critical business questions continue to be answered in the old-fashioned way: based on intuition, gut instinct, and personal experience. In our modern world, however, this is not sufficient anymore and it is essential to supplement the businessperson's gut instinct with science, the science of business analytics. It is said that "money makes the world go around." More than stock trading, the way that it drives the world is through business decisions and actions, including sales and marketing, forecasting and planning, and hiring and attrition. Signal processing applied to those endeavors is the new frontier of business analytics, and it is open for business. We invite you to join the party and help push the frontier of signal processing outward by working on demanding problems motivated by business applications.

From a signals and systems perspective, we will introduce a business as a simplified system with equity and liabilities as inputs, dividends as output, and retained profit and liability repayment as feedback. We will define the financial terminology of signals flowing through the system as well as of figures of merit of system performance known as key financial indicators. We will also describe how those key financial indicators can be controlled by minimizing costs of goods sold, and by effectively using expenditures in marketing. In reality, a business is a highly complex system that cannot be analyzed and optimized holistically; specific components or aspects must be examined separately.

We will focus on specific aspects of cost minimization and effective marketing addressable by various signal processing techniques by exposing the audience to as wide a variety of business analytics problems as time permits. Our aim is to have audience members gain an understanding of the research possibilities that business provides rather than give highly technical descriptions of signal processing algorithms applied to those business problems. Problems covered will come from workforce analytics, salesforce analytics, and market intelligence, and may include dynamic workforce modeling and optimization, talent retention analytics, seller productivity profile estimation, sales recommendation systems, buyer-seller social network analysis, strategic outsourcing management, and other topics. Signal processing topics that may arise include window filters, matrix factorization, quantile regression, Markov chains, stochastic loss networks, logistic regression, and others.

Speaker Biography

Kush R. Varshney received the B.S. degree (magna cum laude) in electrical and computer engineering with honors from Cornell University, Ithaca, NY. He received the S.M. degree and the Ph.D. degree, both in electrical engineering and computer science at the Massachusetts Institute of Technology (MIT), Cambridge.

He is a research staff member in the Business Analytics and Mathematical Sciences Department at the IBM Thomas J. Watson Research Center, Yorktown Heights, NY. While at MIT, he was a research assistant with the Stochastic Systems Group in the Laboratory for Information and Decision Systems and a National Science Foundation Graduate Research Fellow. He has been a visiting student at Laboratoire de Mathématiques Appliquées aux Systèmes at École Centrale, Paris, and an intern at the Systems and Decision Sciences Section, Lawrence Livermore National Laboratory, Livermore, CA, at Sun Microsystems, Burlington, MA, and at Sensis Corporation, DeWitt, NY. His research interests include statistical signal processing, machine learning, image processing, and business analytics.

Dr. Varshney is a member of Eta Kappa Nu, Tau Beta Pi, INFORMS, and IEEE. He received a best student paper travel award at the International Conference on Information Fusion and has received two IBM Research technical accomplishment awards for business impact of outsourcing analytics and for software sales analytics.