Full program with abstracts

ISBIS 2017 Program with Abstracts


Poster abstracts

ISBIS 2017 Poster Abstracts




Wednesday June 7th 2017




0. Welcome Address, Auditorium

Dario Gil, Vice President, Science and Solutions, IBM Research




1.  Keynote Address, Auditorium

Brenda L. Dietrich, IBM Fellow and VP, “Riding Technology Waves: Perspectives and Opportunities for Leveraging Data”




2. ASMBI Invited Session, Room CR3

Org/Chair: Emmanuel Yashchin, IBM Research

• Speaker 1: Christine Anderson-Cook, Los Alamos National Laboratory, ”Graphics to Facilitate Informative Discussion and Team Decision-Making”

• Speaker 2: Jennifer van Mullecom, Virginia Tech, Discussant

• Speaker 3: Tom Donnelly, SAS, Discussant


3. Statistical Models for Complex Data, Room 20-043

Org/Chair: Veronika Rockova, University of Chicago

• Speaker 1: Daniel Zantedeschi, The Ohio State University, ”An Interaction Analysis of Social Media and Traditional Platform Effects in the Consumer Purchasing Funnel”

• Speaker 2: Richard Hahn, University of Chicago, ”Bayesian Causal Forests”

• Speaker 3: Veronika Rockova, University of Chicago, ”Sparse Autoregressive Processes for Dynamic Variable Selection”


4. Recent Advances in Spatio-Temporal Statistical Modeling, Room 20-001

Org/Chair: Huijing Jiang, IBM Research

• Speaker 1: Marco Ferreira, Virginia Tech, ”Dynamic Multiscale Spatiotemporal Models for Poisson Data”

• Speaker 2: Rodrigue Ngueyep Tzoumpe, IBM Research, ”Spatially Fused Time-Varying Lattice Models for Agricultural Management Zoning”

• Speaker 3: Maria Terres, The Climate Corporation, “Spatio-Temporal Data Science at The Climate Corporation”


5. Statistical Applications in the Financial Services and Media Industries, Auditorium

Org/Chair: Aliza Heching, IBM Research

• Speaker 1: Arun Verma, Bloomberg, ”Quantitative Strategies Using Sentiment Classification of Financial News Using Statistical Techniques in Machine Learning”

• Speaker 2: Samer Takriti, Viking Global Inverstors, ”Multi-Factor Models: Risk and Attribution”

• Speaker 3: Rahel Jhirad, Hearst, “Applications of Machine and Deep Learning in Media”




6. Data-Driven Learning in Industrial Settings, Room 20-043

Org/Chair: Yada Zhu, IBM Research

• Speaker 1: Brian Quanz, IBM Research, ”Towards Cognitive Product Data Cleaning in e-Commerce”

• Speaker 2: Yingdong Lu, IBM Research, “Moderate Deviations in Cloud Computing via Stein-Chen Method”

• Speaker 3: Yada Zhu, IBM Research, ”An Efficient Graph Algorithm for Customer Return Prediction in e-Commerce Industry”


7. Statistics in the IT Industries, Auditorium

Org/Chair: David Banks, Duke University

• Speaker 1: Erin LeDell, H2O, ”Towards Scalable Automatic Machine Learning”

• Speaker 2: Timothy Au, Google, ”Random Forests, Decision Trees, and Categorical Predictors: The ‘Absent Levels’ Problem”

• Speaker 3: Xi Chen, LinkedIn, “Scalable Bayesian Modeling and Monitoring of Dynamic Network Flow Data with Online Advertising Applications”


8. Data Science for Decision Support, Room CR3

Org/Chair: Kay See Tan, Memorial Sloan Kettering Cancer Center

• Speaker 1: Nancy Hersh, Independent, ”Using analytics to measure and increase energy savings”

• Speaker 2: Karina Kervin, IBM Research, “Automated External Feature Sensitivity Scoring Tool”

• Speaker 3: Maxwell Shron, Warby Parker, ”Interfacing With Decision Makers Across Warby Parker”




9. Impact Measurement Methodologies, Auditorium

Org/Chair: Marianna Dizik, Google

• Speaker 1: Yuxue Jin, Google, ”Bayesian methods in media mix modeling”

• Speaker 2: Qing Wu, Google, ”Longitudinal approach to measure treatment impact”

• Speaker 3: Yongjian Kang, Google, ”Impact Measurement using SEM”


10. Applications of Data-Driven Decision Making, Room 20-043

Org/Chair: Tahir Ekin, Texas State University

• Speaker 1: Francis Mendez, Texas State University, ”Data-Intensive Time and Motion Studies for Manufacturing Operations”

• Speaker 2: Sitki Gulten, Stockton University, ”High-Frequency Trading in Risk-Averse Portfolio Optimization with Higher-Order Risk Measures”

• Speaker 3: Tahir Ekin, Texas State University, ”Data-driven pre-screening of claims for medical audits”


11. Reliability, CR3

Org/Chair: Kassie Fronczyk, Institute for Defense Analyses

• Speaker 1: James Gilman, North Carolina State University, ”Reliability Assessment of Multiple Component Systems Using Bayesian Hierarchical Models”

• Speaker 2: Saurabh Mishra, International Finance Corporation, World Bank Group, University of Maryland College Park, ”Economic Complexity and Globalization of Services”

• Speaker 3: Vincent Raja Anthonisamy, University of Guyana, ”Reliability Modeling Incorporating Load Share and Frailty”


Thursday June 8th, 2017




1.  Keynote Address, Auditorium

Merlise A. Clyde, Professor of Statistical Science, Duke University, “Bayesian Model Choice: Past, Present, Future”




2. From Samples to Target Populations, Room 20-043

Org/Chair: Chaitra H. Nagaraja, Fordham University

• Speaker 1: Frank Yoon, IBM Watson Health, ”Observational Methods for Health Policy Decision-Making”

• Speaker 2: Elizabeth Tipton, Teachers College, Columbia University, ”Designing randomized trials for making generalizations to policy-relevant populations”

• Speaker 3: Chaitra H. Nagaraja, Fordham University, ”An instrumental variable approach to generalizing experimental results”


3. Statistics Engineering at Facebook, Room CR3

Org/Chair: Daniel Merl, Facebook

• Speaker 1: Ben Letham, Facebook, ”Forecasting at Scale”

• Speaker 2: Brian Karrer, Facebook, ”Bayesian optimization for infrastructure systems”

• Speaker 3: John Myles White, Facebook, ”Challenges of A/B Testing at Facebook”


4. Spatio-Temporal Statistics for Environmental Sciences and Climatology, Auditorium

Org/Chair: Huijing Jiang, IBM Research

• Speaker 1: Candace Berrett, Brigham Young University, ”Bayesian Spatio-Temporal Factor Analysis for Prediction”

• Speaker 2: Taps Maiti, Michigan State University, ”Spectral Radiance in Climate Study”

• Speaker 3: Lloyd Treinish, The Weather Company & IBM Research, ”Coupled Physical and Statistical Models for Renewable Energy Integration”




5. Bayesian Analysis for Large and Complex Data, Room CR3

Org/Chair: Tamara Broderick, MIT

• Speaker 1: Barbara Engelhardt, Princeton University, “Online Patient Monitoring Using Medical Time Series Data”

• Speaker 2: Katherine Heller, Duke University, ”Machine Learning for Improving Healthcare”

• Speaker 3: Tamara Broderick, MIT, ”Approximate Sufficient Statistics for Scalable Bayesian Inference”


6. Machine Learning with Mixed Data Types, Auditorium

Org/Chair: Bonnie Ray, Arena

• Speaker 1: Elizabeth Schifano, University of Connecticut, ”Variable selection for correlated bivariate mixed outcomes using penalized generalized estimating equations”

• Speaker 2: Marianthi Markatou, Dept. of Biostatistics, University at Buffalo, ”A Semiparametric Method for Clustering Mixed Data”

• Speaker 3: Naveen Naidu Narisetty, University of Illinois at Urbana-Champaign, “Scalable Computation with Skinny Gibbs Sampler for High Dimensional Bayesian Models”


7. Web Applications in Statistics and Machine Learning, Room 20-043

Org/Chair: Cheryl Flynn, AT&T Labs Research

• Speaker 1: Shawndra Hill, Microsoft Research, ”Television and Digital Advertising: Second Screen Response and Coordination with Sponsored Search”

• Speaker 2: Roman Vaculin, IBM Research, “Blockchain and Ledger Analytics”

• Speaker 3: Zhenming Liu, College of William and Mary, “From Which World Is Your Graph?”

• Speaker 4: Cheryl Flynn, AT&T Labs, ”Deconstructing Domain Names to Reveal Latent Topics”




8. Data Tales from Industry, Auditorium

Org/Chair: Claudia Perlich, Dstillery

• Speaker 1: Jon Krohn, Untapt, ”Match Making for Tech Jobs”

• Speaker 2: Peter E Lenz, Dstillery, ”You are probabilistically here: Lessons from working with mobile device generated geodata”

• Speaker 3: Ravi Shroff, New York University, ”Improving Police Stop Efficiency in New York City”


9. Statistical Methods in Medicine, Room CR3

Org/Chair: Rebecca Yates Coley, Group Health Research Institute

• Speaker 1: Lindsay Berry, Duke University, ”Simulation Controlled Seamless Phase II/III Clinical Trials”

• Speaker 2: Sandy Griffith, Flatiron Health, ”Unlocking real-world oncology data using electronic health records”

• Speaker 3: Rebecca Yates Coley, Kaiser Permanente Washington Health Research Institute, ”A Data Science Framework for Learning Health Systems”


10. Data Driven Semi-Parametric Models, 20-043

Org/Chair: Emre Barut, George Washington University

• Speaker 1: Harlan Harris, WeWork, ”Predicting Class Size for Fun and Profit with Semi-parametric Regression Models”

• Speaker 2: Youngdeok Hwang, IBM Research, ”Bayesian Pollution Source Identification via an Inverse Physics Model”

• Speaker 3: Emre Barut, George Washington University, ”An Efficient Method for Parameter Estimation Under Model Contamination”


Friday June 9th, 2017




1.  Keynote Address, Auditorium

Moorea Brega, Sr. Director, Data Science, Premise Data, “When Less is More: Adaptive Data Collection for Economic Indicators”




2. Panel Discussion: Succeeding as an Early-Career Data Scientist in Today’s Industry, Room 20-001

Org/Chair: Grant Weller, Savvysherpa

• Speaker 1: Mattia Ciollaro, Spreemo Health, Panel discussant 1

• Speaker 2: Reka Daniel-Weiner, Dstillery, Panel discussant 2

• Speaker 3: Ryan Roundy, Oracle, Panel discussant 3


3. Novel Hierarchical Bayesian Approaches for Business and Government Applications, Auditorium

Org/Chair: Nalini Ravishanker, University of Connecticut

• Speaker 1: Sujit K. Ghosh, NCSU and SAMSI, “Regression and Reliability Models for Predicting Customer Churning”

• Speaker 2: Scott Holan, University of Missouri and US Census Bureau, ”Multivariate spatio-temporal survey fusion with application to the American Community Survey and Local Area Unemployment Statistics”

• Speaker 3: Nalini Ravishanker, University of Connecticut, “Dynamic Models for Multivariate Times Series of Counts”


4. New Statistical Methods for Stochastic Volatility and Functional Data in Finance and Engineering, Room 20-043

Org/Chair: Mengyang Gu, Johns Hopkins University

• Speaker 1: Jian Zou, Worcester Polytechnic Institute, ”Efficient Portfolio Allocation with Sparse Volatility Estimation for High-Frequency Financial Data”

• Speaker 2: Xinyi Xu, Ohio State University, ”Shape-constrained Semiparametric Additive Stochastic Volatility Models”

• Speaker 3: Mengyang Gu, Johns Hopkins University, ”Robust Gaussian Stochastic Process”


5. Analytics Promoting Social Good: Money Access, Housing and Health, Room CR3

Org/Chair: Marian Farah, The Climate Corporation

• Speaker 1: Ankur Gupta, Premise Data, ”Route-based data collection to determine mobile money access in Africa”

• Speaker 2: Ryan Brady, Apteligent, ”Risk factors for eviction in rapidly developing cities”

• Speaker 3: Kush Varshney, IBM Research, ”A Bayesian Approach for Predicting Neotropical Primate Reservoirs of Zika Virus”