Wednesday November 16, 12:00 – 2:00 pm
A Framework for Checking and Tuning Up Data Graphics: Lessons from 10+ Years of BloggingOver 10 years ago, I started the Junk Charts blog, the first website dedicated to data visualization criticism. The blog is widely regarded as a key resource for practitioners in the field, containing discussion of many hundreds of charts and graphics that have appeared in the popular press. I will discuss the key components of the Trifecta Checkup, the general framework for assessing the strengths and weaknesses of dataviz products. The framework draws connections between our emotional response to data visualization and the specific choices made by designers. I will demonstrate how to use the Trifecta framework to tune up your own charts. Finally, I offer some tips for those interested in starting a blog.
Wednesday October 26, 12:00 – 2:00 pm
Business Intelligence and Predictive Analytics in the Healthcare EnvironmentThe healthcare industry is fortunate to have extraordinary database resources, originally intended for social planning purposes, which can now be deployed for Business Intelligence and Population Health purposes. However the complexity and quality of these data repositories must be thoroughly understood and effectively managed to garner the information and intelligence so desperately needed by the evolving regulatory, economic and competitive environment. While privacy and confidentiality issues abound in this data environment (HIPAA), the opportunity to understand Community Based Needs and the trends towards a new economic model require all of the expertise of the Data Science field. The infrastructure and quality of massive repositories of structured and unstructured data continue to explode along with the constant evolution of the classifications and codes employed. The capacity to mine the data is equally as dynamic as the tools to evaluate trends and seasonal patterns. Additionally the expertise is evolving to provide longitudinal linkages to understand the progression of conditions and diseases as a part of clinical decision support and the understanding of 'Risk'. We will explore these issues in the context of an industry in fundamental transformation.
Wednesday September 21, 12:00 – 2:00 pm
Machine Learning and Artificial Intelligence applied to a Pharmaceutical Marketing Problem
Machine Learning and Artificial Intelligence have increasingly been used as innovative ways to extract meaningful value from data. This presentation will describe how those techniques were applied to generate a set of business rules and a predictive model for a pharma company to increase their physician client base’s propensity to prescribe their drug to their patients as opposed to the competition’s drug(s). The result is a personalized recommendation for each physician, with accompanying auditable rules to audit results and validate success.
Wednesday June 15 , 2016
The New York City Police Department's Domain Awareness System
Finalist Presentation in the INFORMS 2016 Franz Edelman Award CompetitionThe New York City Police Department (NYPD), the largest state or local police force in the United States, is charged with securing New York City from crime and terrorism. The NYPD's Domain Awareness System (DAS) is a citywide network of sensors, databases, devices, software, and infrastructure that informs decision making by delivering analytics and tailored information to officers' smartphones and precinct desktops. DAS development began in earnest in 2008; since then, the NYPD has used the system to employ a unique combination of analytics and information technology, including pattern recognition, machine learning, and data visualization.
WednesdayDecember 14, 12:00 – 2:00 pm
Chief Scientist, Distillery
Data Scientist, Distillery
Bridging the Gap: Connecting online behaviors to offline purchases
Despite many advances, digital advertising has been struggling to address the most basic advertising question: can we measure and draw conclusions about the effectiveness of advertising on consumer behavior? Being fraught with challenges like questionable viewability, attribution and fraud in a pure online setting, consumer packaged Goods (CPG) companies have additional challenges gauging advertising effectiveness: Purchases do not happen online and linking third parties scanner data from grocery stores introduces further scale limitations and match uncertainty. Here we present a case study showing that it is possible to seamlessly digitize, access and integrate real world physical behaviors into a large-scale digital experiment. We show that using using causal analytical methods combined with machine learning, sufficient statistical power can be achieved to measure significant and fine-grained effects. The resulting insights cannot only be used to assess the efficacy and economic viability of running display advertising campaigns, but also to optimize creatives for each prospective customer -- allowing marketers to not only to find the most promising prospects, but also to match them to the right creative.
For reservation assistance write to our gmail address or call John Waldes (212) 362-0302
INFORMS NY Metro is actively seeking speakers and topics. If you have a data analytic or business oriented topic please contact Jack Theurer or Leon Schwartz.
INFORMS NY is the New York City Chapter of the Institute for Operations Research and Management Science.
Part-time faculty position for Operations Management at Zicklin School of Business, Baruch College, City University of New York (CUNY)