U.S. Department of Energy

Pacific Northwest National Laboratory

News

2017

Bramer accepted to Data Science Innovation Lab

Lisa Bramer has been accepted to the 2017 Data Science Innovation Lab: Microbiome. The intensive five-day workshop in intended to facilitate the development of new teams of early-career biomedical and quantitative investigators who generate multidisciplinary cooperative research programs through a real-time and iterative mentoring process.

Greaves Selected for Editorial Board

Mark Greaves has been selected as a guest editor for Semantic Web Journal’s Special Issue on Stream Reasoning.

Submissions are currently being accepted for the issue, and are due by September 17.

Volkova Inaugural Member of ACM Future of Computing Academy

Congratulations to Svitlana Volkova on her selection as a member of the ACM Future of Computing Academy (FCA). Svitlana was accepted, with 45 of your colleagues, from a pool of 300 applicants to form the inaugural class of the FCA.

Cottam paper accepted to FOAP Workshop

Joseph Cottam’s position paper, “Interaction Response at the Network's Edge,” has been accepted to the FuturePlatform Workshop.

Sponsored by Future Online Analysis Platform, the workshop was held April 4-5 in Arlington, VA.

Volkova on Program Committee for SocialNLP 2017

Svitlana Volkova is on the program committee for the Workshop on SocialNLP 2017 – the 5th International Workshop on Natural Language Processing for Social Media. The workshop was held on April 3 in Valencia, Spain.

Nichols featured in STEMed Video

“Women are changing the world with their careers in science and technology,” stated Nicole Nichols, the feature of a video about women in STEM. Nicole speaks about her path to hear current role, and obsticles overcome to achieve her goals.

youtu.be/1QPBn4yCd4Q 

Two PNNL Papers Accepted at ICWSM-17

PNNL staff had two papers accepted to the 11th International Conference on Web and Social Media (ICWSM-17). Svitlana Volkova and Eric Bell authored “Identifying Effective Signals to Predict Deleted and Suspended Accounts on Twitter across Languages,” while Ian Stewart, Dustin Arendt, Eric Bell, and Svitlana wrote “Measuring, Predicting and Visualizing Short-Term Change in Word Representation and Usage in VKontakte Social Network.” http://www.icwsm.org/2017/index.php

Research featured on Cover

Research by Leslie Blaha et al is featured on the cover on Eye Tracking and Visualization Foundations, Techniques, and Applications. The book is a collection of papers from the  First Workshop on Eye Tracking and Visualization, part of IEEE VIS Conference 2015. 

http://www.springer.com/us/book/9783319470238 

ACL 2017 Program Committee

Svitlana Volkova has been selected as the area chair on Social Media for the Association for Computational Linguistics 55th Annual Meeting – ACL 2017. She is also co-organizing a workshop on Natural Language Processing and Computational Social Science at the workshop.

Automated Crystal Growth Analysis Techniques

Scientific image analysis is a labor intensive manual process exacerbate by growth in rate and volume of data. PNNL research has demonstrated that video shot boundary detection can automatically detect frames where a change in the image occurs, allowing for automated segmentation of a digital stream for further analysis. 

 

2016

Blaha Selected for Society for Mathematical Psychology Board

The Society for Mathematical Psychology board has voted in Leslie Blaha for a four-year appointment as the Secretary/Treasurer. SMP is the premier society promoting the advancement and communication of mathematical psychology (and related disciplines) research. Leslie additionally sits on the Women of Math Psych Organizing Committee. www.mathpsych.org

Choudhury Invited Speaker at Chesapeake Large-Scale Analytics Conference

Sutanay Choudhury was invited to speak at the Chesapeake Large-Scale Analytics Conference, organized by the Association for High Speed Computing. Sutanay’s talk, Knowledge Graphs - From Data to Insight, was held on October 25. http://www.clsac.org/clsac16.html

Research Presented at IEEE VIS 2016

Lindsey Franklin, with Kris Cook, R. Jordan Crouser (Smith College), and Alex Endert (Georgia Institute of Technology) presented “Toward Theoretical Techniques for Measuring the Use of Human Effort in Visual Analytic Systems” during the VAST: Managing Visual Analytics Process track at IEEE VIS 2016.  

Arendt, Pirrung IEEE VGTC Best Poster

Dustin Arendt and Meg Pirrung’s poster was one of two winning submissions in the IEEE Visualization and Graphics Technical Committee (VGTC) International Data-Visualization Contest. The annual competition recognizes outstanding design in visualization and visual analytics.

Volkova Speaks on Social Media Users at GHC

Svitlana Volkova presented “Predicting User Demographics, Emotions and Opinions in Social Networks” at the recent Grace Hopper Conference. GHC is the world’s largest conference of women technologists, boasting more than 15,000 attendees.  http://ghc.anitaborg.org/

Best Paper at ETVIS 2016

Leslie Blaha, Dustin Arendt, and Tim Balint (former PNNL Intern) were awarded best paper award for “Storyline Visualizations of Eye Tracking of Movie Viewing” at ETVIS 2016, The Workshop on Eye Tracking and Visualization. ETVIS is held in conjunction with IEEE VIS.

Remembering Important Things

“Remembering the Important Things: Semantic Importance in Stream Reasoning” was presented at the Stream Reasoning Workshop 2016, Kobe, Japan. The short paper was authored by Rui Yan, Mark Greaves, Wil Smith, and Deborah McGuinness.  

Cover Story

Dustin Arendt, with Russ Burtner, provided the cover art IEEE Computer Graphics and Applications September/October 2016 issue, which also featured a story on Dustin’s research and was distributed to participants at the conference. Arendt's research was also featured in an article in the publication.

Vishnu Presents at ICPP

Abhinav Vishnu’s paper “Fault Tolerant Support Vector Machines” was accepted and presented at the International Conference on Parallel Processing (ICPP), August 16-19, 2016. http://icpp2016.cs.wcupa.edu/

Hodas "Does the Impossible" at NCPW15

Nathan Hodas presented "Doing the impossible: Why neural nets can be trained at all"  at the 15th Neural Computation and Psychology Workshop, Philadelphia, PA. 

Modeling Temporal Dependence at JSM 2016

“Modeling Temporal Dependence to Improve Learning Algorithms for Streaming Data” was presented in August at JSM 2016. The paper, authored by AIM’s Lisa Bramer, Bryan Stanfill, and Sarah Reehl with Iowa State University’s Maggie Johnson and Petrutza Caragea, is focused on the use of statistical models incorporating temporal dependence to improve upon current state-of-the-art learning algorithms.

Blaha presents on HCI at MP2016

"Functional Analysis of Movement Behavior for Human-Computer Interaction" was recently presented by Leslie Blaha at the Society for Mathematical Psychology in New Brunswick, NJ. 

Materials Research Society Meeting welcomes Stevens

"Active Learning, Convolutional Factor Analysis, and Compressive Sensing for STEM Acquisition of Beam Sensitive Specimens" was presented at the 2015 Materials Research Society Fall Meeting in Boston, MA by Andrew Stevens. See more about the conference at http://www.mrs.org/fall2015/    

Compressive STEM-EELS Research at M&M 2016

Andrew Stevens recently presented "Compressive STEM-EELS" at the Microscopy and Microanalysis 2016 in Columbus, OH.

Compressive Sensing in Microscopy at ASCI 2016

Andrew Stevens presented ' Compressive Sensing in Microscopy: A Tutorial"  at the fourth annual Advances in Structural & Chemical Imaging (ASCI) Workshop May 18-20, 2016 in conjunction with Boise State University in Boise, ID.

Contributing to OpenKE

Sutanay Choudhury, with his AIM’s NOUS project team, has been contributing to the Laboratory for Analytics Science’s OpenKE project for the past year, which includes developing and illustrating knowledge graphs. In May, Sutanay presented an overview of PNNL’s contributions to LAS during the laboratory’s first-ever two-week Focused Discovery Activity (FDA) that included 40 partners from the Intelligence Community, academia, and industry.

Griswold presents at Nuclear Magnetic Resonance Conference

Richard Griswold attended the 57th Experimental Nuclear Magnetic Resonance Conference in Pittsburgh, PA. He presented "Compression Algorithm Analysis of 1D NMR Data, Towards Automatic Detection and Characterization of Events," about using compression algorithm metadata to detect events in a stream of NMR spectra. Learn more at http://enc-conference.org/

Reasoning Framework presented at LDOW2016

Wil Smith’s paper has been accepted to the Linked Data on the Web (LDOW2016) workshop. He will be presenting Towards A Cache-Enabled, Order-Aware, Ontology-Based Stream Reasoning Framework with co-authors Rui YanBrenda Praggastis, and Deborah L. McGuinness. The workshop takes place on April 12 in Montreal, Canada. http://events.linkeddata.org/ldow2016/papers/LDOW2016_paper_13.pdf 

TensorFlow for the extreme scale

A team of PNNL researchers has now extended TensorFlow to run on extreme scale systems. With a planned release this spring, users can solve very large problems on supercomputers and cloud systems. PNNL will also have a release integrated with Machine Learning Toolkit for Extreme Scale (MaTEx), a collection of high performance parallel machine learning and data mining algorithms is planned for late spring. Learn more at http://goo.gl/lOucUQ.

AI discussion at MIT Enterprise Forum

Mark Greaves recently spoke at the MIT Enterprise Forum Northwest on the Future of Artificial Intelligence. Mark led a discussion about machine learning, deep learning, and general AI. http://goo.gl/fxeQS4 

Welcome Leslie Blaha

The AIM team welcomes Leslie Blaha! Leslie’s research interests include Smart Adaptive Interfaces that will transform the way we use cognitive models to support and enhance human decision making.

AIM featured in ASCR Discovery

AIM was recently featured in ASCR Discovery, and online publication produced by the Office of Advanced Scientific Computing Research (ASCR) in the Department of Energy Office of Science. The article, “Moving target,” discusses how AIM is devising new ways computer algorithms can help researchers interact with streaming data in real time.

PNNL at SXSW

PNNL’s Jenn Hodas was the first-ever national laboratory presenter at the prestigious South-by-Southwest (SXSW) event in Austin, Texas. Hodas, a commercialization manager, presented in the Government & Policy track, which included a keynote address from President Barack Obama. As part of her presentation, Hodas showed this video. Check out the video on YouTube. https://www.youtube.com/watch?v=bZ1mqU_z23Q

Smith at Seattle DAML Meetup

Wil Smith was a recent speaker at the Seattle DAML Meetup: Steaming Hypothesis Reasoning from PNNL. In his talk, he addressed streaming data challenges in sampling rates, cache maintenance, deductive reasoning, and the surrounding Semantic Web framework.   

Teuton presents at CoDA 2016

Jeremy Teuton recently attended CoDA 2016 -- the Conference on Data Analytics which is focused on exploring data-focused research across the Department of Energy. Jeremy presented the poster "Analysis of Streaming Data Using Compression Algorithm Metadata," and networked with researchers from around the DOE complex. 

2015

Stevens presents at 2015 Materials Research Society Meeting

Andrew Stevens presented "Bayesian Machine Learning for Compressive (S)TEM Image/Video Acquisition" at the 2015 Materials Research Society Fall Meeting in Boston, MA.   

Arendt wins top prize in international data visualization contest

The IEEE Visualization and Graphics Technical Committee Visualization Pioneers Group International Data-Visualization Contest was held recently at IEEE VIS 2015. Dustin Arendt, along with Yanina Levitskaia (University of Washington), received the Overall Excellence award for their poster “Telling the Story: Gender Discrepancies in Engineering and Computer Science.”

Vishnu and Agarwal present at IEEE Cluster 2015

Abhinav Vishnu and Khushbu Agarwal, along with their collaborators, recently presented their research involving examining extreme-scale machine learning and data mining algorithms scaling data to large supercomputers.

Open source software for the extreme scale

Abhinav Vishnu presented “MaTEx: Machine Learning Toolkit on Extreme Scale” at the Supercomputing 2015 Conference in Austin, TX. View the 20-minute presentation on YouTube: https://youtu.be/YZfvKHQNhT8

SHyRe Highlighed at Smart Data Conference

Wil Smith presented "The PNNL Streaming HYpothesis REasoning (SHyRe) Use Case" recently at the Smart Data Conference in San Jose, CA. http://smartdata2015.dataversity.net/ 

Bramer presents at JSM 2015

Lisa Bramer presented "A meta-analysis approach to exploratory data analysis and feature exploration for complex ‘omics data through computational statistics" at the Joint Statistical Meetings in Seattle, WA. https://www.amstat.org/meetings/jsm/2015/

MaTEx version 0.2 now available

MaTEx version 0.2 combats the issue of scale related to cluster computing. Abhinav Vishnu and team have released The Machine Learning Toolkit for Extreme Scale (MaTEx), an open source software that is available across different platforms, targeted for desktops, supercomputers and cloud computing systems. Version 0.2 is available for download at http://hpc.pnl.gov/matex/

Networking, learning, stimulating new ideas.

Lisa Bramer recently attended the 10th Annual Women in Machine Learning Workshop in Montreal, Canada http://www.wimlworkshop.org. The technical workshop provides an opportunity for females working in the field to network, exchange ideas, and learn from each other.

Vishnu and Agarwal present at IEEE

Abhinav Vishnu and Khushbu Agarwal, along with their collaborators, recently presented their research involving examining extreme-scale machine learning and data mining algorithms scaling data to large supercomputers at IEEE Cluster 2015. "Large Scale Frequent Pattern Mining using MPI One-Sided Model" was presented during the Machine Learning and Data Mining session.

Teuton dazzles with video

“Compression Algorithm Analysis of In-Situ (S)TEM Video: Towards Automatic Event Detection and Characterization" was presented by Jeremy Teuton at the Microscopy and Microanalysis Conference, Portland, OR.

Bramer travels to Vegas for WORLDCOMP’15

Lisa Bramer presented "A Machine Learning Approach for Business Intelligence Analysis Using Commercial Shipping Transaction Data" at WORLDCOMP'15 - the International Conference on Data Mining, in Las Vegas, NV. Visit the website for more information.

Heredia-Langner takes on DMIN 2015

Alejandro Heredia-Langner was a speaker at DMIN'15: The 11th International Conference on Data Mining, Las Vegas, NV on July 27, 2015. He presented "Selecting a Classification Ensemble and Detecting Process Drift in an Evolving Data Stream." Learn more... 

Greaves contributes to National and Transnational Security report

The American Association for the Advancement of Science (AAAS), Federal Bureau of Investigation (FBI), and United Nations Interregional Crime and Justice Research Institute recently completed a year-long study of the benefits and risks of big data, and the related national security implications. 

Kleese van Dam invited as a keynote speaker

Kerstin Kleese van Dam has been invited to give a keynote speech at the Data Science Innovation Summit, in San Diego, CA. She will be presenting on "How to use Streaming Data for Real Time Discovery and Decision Taking."

Kleese van Dam travels to big data workshop

Kerstin Kleese van Dam will be attending the Big Data Strategic Initiative Workshop organized by the Network and Information Technology Research and Development Program's (NITRD) Big Data Senior Steering Group.

Pages

| Pacific Northwest National Laboratory