UFII Annual Symposium 2017


                               UFII 3rd Annual Symposium 2017

UFII Symposium

The University of Florida Informatics Institute hosted its 3rd Annual Symposium on Thursday, March 16, 2017, at the Reitz Union. Students, researchers, faculty, and industry professionals from across the nation joined together to interact, share research, and collaborate. Focusing on the latest trends in informatics, and our Mission of cross-discipline collaborative research, we had invited guest speakers from across the campus and outside the University to present on  topics, including election data, text mining, network science, and machine learning.


The program included keynote speakers across a diverse range of scientific fields, with a poster session showcasing UFII Fellows and SEED funded scientists.


Dr. George Michailidis, Director of UF Informatics Institute and Professor of Statistics and CISE, University of Florida
“Introduction and Update on UFII”

Gordon Wilson, DSI President, University of Florida
“UF DSI: Bringing Data Science to the Broader UF Community”

Dr. Kathleen M. Carley, Professor of Societal Computing and Director of the Center for Computational Analysis of Social and Organizational Systems, Carnegie Mellon University
“Wisdom or Ignorance: The heedful manipulation of networks”
Abstract: It is generally well accepted that your position in the social network affects your ability to get information. However, we often overlook the fact that you can also use your position to impact what information others receive. In this talk, these ideas are brought together to ask, how do the network positions of those with whom you interact, influence you? This issue is explored using high dimensional network data. Case examples ranging from the Arab Spring to Support for Autistic children are shown to illustrate two basic principles. First – heedful network coordination among alters can effect better outcomes for ego. Second, heedful network manipulation by alters can create communities of ignorance. Data is drawn from interview, survey, and social media data. Issues related to data sampling and big data are discussed. The results suggest that where networks are concerned, the wisdom of the crowd can be manipulated.

Dr. Bonnie Dorr, Associate Director and Senior Research Scientist, Florida Institute for  Humans & Machine Cognition, Ocala
“Natural Language and Data Science: Events in Cyberspace and Evaluation-Driven Research”
Abstract: Natural language researchers and data scientists apply a multitude of techniques for analysis and extraction of knowledge from potentially massive data. Language processing of social media to discover possible cyber threats is one example of a problem that falls under the heading of data science. Natural language analysis of online media to extract goals and intentions may lead to predictions of when an event may occur in cyberspace. This talk will present concepts behind the implementation of techniques for processing large amounts of formal and informal textual data to build necessary evidence for predictive analytics. The talk includes a discussion of how progress in data-science research is enabled and enriched through a cross-field examination of problems, solutions, measures, and evaluation paradigms (in fields as diverse as traffic, finance, sports, and language), thus leading to the discovery of solutions that otherwise would not have been apparent within a given field.

Dr. Robert Guralnick, Associate Curator of Biodiversity Informatics, University of Florida [SEED funding]
“Informatics for Global Mapping and Monitoring of Biodiversity”
Abstract: I will be presenting two separate topics related to informatics for biodiversity. The first covers how to assemble large-scale trait data from digitized and mobilized specimen records, and the value of that enterprise for data intensive science. I will discuss methods for doing this in a semantic and graph database framework that provides the most useful framework for data integration and fusion. I will also touch on work related to Map of Life, which attempt to assemble and model species distributions for tens to hundreds of thousands of species and the importance of this effort in policy frameworks related to biodiversity monitoring.

Dr. Georgios Giannakis, ADC Chair in Wireless Telecommunications and McKnight Presidential Chair in ECE, University of Minnesota
“Adaptive Sketching and Validation for Learning from Big Data”
Abstract: We live in an era of data deluge. Pervasive sensors collect massive amounts of information on every bit of our lives, churning out enormous streams of raw data in various formats. Mining information from unprecedented volumes of data promises to limit the spread of epidemics and diseases, identify trends in financial markets, learn the dynamics of emergent social-computational systems, and also protect critical infrastructure including the smart grid and the Internet’s backbone network. While Big Data can be definitely perceived as a big blessing, big challenges also arise with large-scale datasets. This talk will put forth novel algorithms and present analysis of their performance in extracting computationally affordable yet informative subsets of massive datasets. Extraction will effected through innovative tools, namely adaptive censoring, random subset sampling (a.k.a. sketching), and validation. The impact of these tools will be demonstrated in tasks as basic as regression and clustering of high-dimensional, large-scale, and dynamic datasets emerging with power and social network applications.

Dr. Xiaolin “Andy” Li, Associate Professor of Electrical Engineering and Computer Engineering, University of Florida [SEED funding]
“NSF I/UCRC for Big Learning: Creating Intelligence via Large-scale Deep Learning”
Abstract: This talk will give an overview of the renaissance of artificial intelligence driven by deep learning, deep reinforcement learning, and general machine learning. Representative models and capabilities of deep learning will be introduced. Representative projects will be highlighted and elaborated, including DeepCloud intelligent collaborative platforms, GatorPilot self-driving ecosystems, DeepHealth for precision medicine, sepsis, cancer and genomics plus other projects for cybersecurity, science, and business.

NSF I/UCRC Center for Big Learning (CBL) is under the auspices of the industry & University Cooperative Research Center Program of National Science Foundation. With the vision of creating intelligence and leveraging collective wisdom from academia, industry, and governments, the CBL consortium focuses on large-scale deep learning, intelligent platforms, and DL-enabled big data applications in a broad spectrum of disciplines. More information can be found at http://nsfcbl.org.

Dr. Michael McDonald, Associate Professor of Political Science, University of Florida [SEED funding]
“What Can Big Election Data Tell Us About the 2016 Presidential Election?”
Abstract: On election night the world was stunned as Donald Trump was elected president. No pollster or election prognosticator had predicted with confidence that Trump would lose the popular vote by 2.8 million votes, but win three key battleground states in the Electoral College by a combined 77,000 votes. However, there were pre-election indicators lurking in Big Election Data that Clinton’s victory was not as secure as the polling suggested. Here I review some of these telltale signs leading into the election, and explore some of what we’ve learned in the aftermath to understand how Trump was elected president.

Dr. Liang Mao, Associate Professor of Geography, University of Florida [SEED funding]
“Developing temporally comparable high resolution rurality maps for social and health sciences”
Abstract: The concept of ‘rurality’ has been widely used by policy makers and governments at all levels to address social, economic, environmental, educational and health issues, such as health disparities, social isolation, and aging problems. Although many efforts have been devoted to distinguishing rural from urban areas (e.g., by the US Census and USDA), current definitions of ‘rural’ are oriented to large statistical areas and temporally incomparable. As a result, these definitions are not suitable for longitudinal studies that are common in health and social sciences. Furthermore, rural effects on health/social outcomes can only be studied at spatially aggregate levels, which are subject to ecological fallacy. To fill this gap, we proposed to fine-grained cell as spatial unit to define rurality, and developed a new rurality map of Florida as a 600-meter grid for three comparable years, i.e., 1990, 2000, and 2010. For each cell location, a continuous rurality degree was computed by combining local demographic, socio-economic, and settlement factors, scored from criteria consistent over the three time periods. As a case study in health, we applied the new rurality map to understand the effects of rural-urban difference on late-stage cancer diagnosis in Florida, which remain a debate in the literature. The results demonstrated that the cell-based rurality can reveal information that would be otherwise clouded by conventional rural-urban classifications. This research highlights an impressing need of developing fine grained rural-urban classifications by federal and state authorities.


Varsha Sundaresan, Ying Li, Benedetto DiCiaccio, Victor T Lin, Lei Zhou
UF Department of Molecular Genetics and Microbiology; UF Health Cancer Center; UF Genetics Institute
“A comparative genomics approach to understanding the control of cell context dependent P53 binding”

Mi Jung Lee, Sergio Romero, Ann L. Gruber-Baldini, Craig Velozo, Lisa M. Shulman
UF Department of Occupational Therapy; Center of Innovation on Disability and Rehabilitation Research, North Florida/South Georgia Veterans Health System; University of Maryland Medical Center; Medical University of South Carolina
“Assessment of a PROMIS Measure of Self-Efficacy for Managing Emotions”

Brian J. Stucky, Ramona L. Walls, John Deck, Robert P. Guralnick
UF Florida Museum of Natural History; CyVerse, University of Arizona; University of California, Berkeley

“OntoPilot: New software to simplify and accelerate ontology development and deployment in automated reasoning pipelines (https://github.com/stuckyb/ontopilot)”

Carl Klarner
UF Department of Political Science
“Measuring Voting Eligible Population in U.S. Counties: 1970-2015”

Hunter R. Merrill, Xueying Tang, Nikolay Bliznyuk
UF Informatics Institute; UF Department of Agricultural and Biological Engineering; UF Department of Statistics
“Spike-and-Slab Variable Selection for Spatio-Temporal Semiparametric Regression Models”

Tomas Bustamante, Robert P. Guralnick, Douglas E. Soltis, Pamela S. Soltis, Jamie Ellis
UF Department of Entomology and Nematology; UF Florida Museum of Natural History; UF Department of Biology; UF Genetics Institute
“Creating an Extensible Crowdsourcing Platform for Collecting Data from Digitized Specimens”

Arsenios Tsokas, Anton Kocheturov, Farnaz Babaie, Petar Momcilovic, Panos Pardalos, Azra Bihorac
Department of Industrial and Systems Engineering; UF College of Medicine – Department of Anesthesiology
“Mining Predictive Temporal Patterns from Intra-Surgical Time Series”

Jugpreet Singh, Mehul Bhakta, Melanie J. Correll, Christopher Hwang, Salvador Gezan, Pepe C. Michelangeli, James W. Jones, Kenneth J. Boote, Eduardo C. Vallejos
UF Department of Horticultural Sciences; UF Department of Agriculture and Biological Engineering; UF School of Forest Resources and Conservation; UF Department of Agronomy
“An Integration of Genomics and Crop Growth Models for Predicting Phenotypes in Changing Climate Conditions”

Clint P. George, Sahba Akhavan Niaki, George Michailidis, Carole R. Beal
UF Department of Statistics; UF Informatics Institute; UF College of Education

The Impact of an Online Tutoring Program for Algebra Readiness on Mathematics Achievements”

Dipsikha Debnath, James S. Gainer, Can Kilic, Doojin Kim, Konstantin T. Matchev, Yuan-Pao Yang
University of Florida; University of Hawaii; University of Texas, Austin; CERN
“Detecting kinematic boundaries in phase space and particle mass measurements”

Xianqi Li, Yunmei Chen
UF Department of Mathematics

A New Backtracking Strategy for Accelerated ADMM”

Monami Banerjee
UF Department of Computer & Information Science and Engineering
“Nonlinear Regression for Manifold-valued Data with Applications”

Alexander Kirpich, George Michailidis, Lauren M. McIntyre
UF Informatics Institute; UF Department of Molecular Genetics & Microbiology; UF Genetics Institute; UF Southeast Center for Integrated Metabolomics; UF Department of Statistics

Variable Selection in Untargeted Metabolomics Data Analysis”

Ragnhildur I. Bjarnadottir, Robert J. Lucero
UF College of Nursing
“Exploring Fall Risk and Prevention in Nurses’ Narrative Notes”

Ragnhildur I. Bjarnadottir, Robert J. Lucero, Yingwei Yeo, Gail Keenan
UF College of Nursing
Facilitating use of NANDA terminology in palliative care: A text mining study”

Wenche Wang, Fan Li
UF Department of Economics
Scores vs. Stars: A regression Discontinuity Study of Online Consumer Review”

Xiaolin ‘Andy’ Li, Ana Conesa, Yanjun Li, Samskruthi R. Padigepati, Xiaoyong Yuan
UF Large-scale Intelligent Systems Laboratory; NSF I/UCRC Center for Big Learning
“Upfolding: End-to-End Prediction of Human Protein in Novo Structures”

Xiaolin ‘Andy’ Li, Pan He, Qile Zhu
UF Large-scale Intelligent Systems Laboratory; NSF I/UCRC Center for Big Learning
“DeepScene Text: End-to-End Scene Text Detection and Recognition”

Xiaolin ‘Andy’ Li, Lily Elefteriadou, Siva Srinivasan, Rahul Sengupta, Jiyao Guo, Pan He, Xiyao Ma, Shandy Sulen, Ryan Berndt
UF Li Lab; UF Transportation Institute; NSF I/UCRC Center for Big Learning
“GatorPilot: Spatiotemporal Memories for Self-Driving Safety via Deep Reinforcement Learning”

Xiaolin ‘Andy’ Li, Daniela Olivera, Andre Gregio, Xiaoyong Yuan, Qile Zhu, Pan He, Aokun Chen 
University of Florida; Federal University of Parana; NSF I/UCRC Center for Big Learning
“DeepMalware: Deep Models and Mechanism for Malware Detection and Defense”

Xiaolin ‘Andy’ Li, Ye Luo, Faming Liang
UF Large-scale Intelligent Systems Laboratory; NSF I/UCRC Center for Big Learning
“DeepSelection: Nonparametric Variable Selection via Penalized Deep Learning”


Our previous symposia have featured posters, presenters and attendees from animal science, biology, biochemistry, computer science, ecology, education, electrical engineering, health sciences, mathematics, and political science. We have hosted speakers from industry and other national groups including National Ecology Observatory Network (NEON) and Defense Advanced Research Projects Agency (DARPA). Our goal is to reach out and include a population that is equally diverse. To learn more, visit the 2015 and 2016 Symposiums.