News

Data Science Institute Supports UA Grad Research

Monday, March 23, 2020

 

Ariyan Zarei processes and analyzes aerial and ground images of crops in order to estimate their different agricultural phenotypes. Marina Kisley infers patterns from astronomical data to enhance our understanding of the universe. And Artin Majdi helps efforts to detect valley fever by exploring chest X-ray images via deep learning, a subset of machine learning, and introduces a non-invasive method for measuring Skin Conductance Response (SCR), a phenomenon that can reveal emotion.

Learn more about these projects 

 

Machine Learning Literacy Workshop held Feb 8, 2020

Friday, March 6, 2020
Image of MLLP Workshop Instructors and participants

 

The Machine Learning Literacy Project (MLLP) was a one-day workshop held in ENR2 at the University of Arizona on Saturday February 8, 2020. The workshop came about to address the serious need of preparing students for the modern Machine Learning-literate workforce.  Due to the amazing things that can now be done using ML, the demand for its use in every discipline from Science and Engineering, to Law, to the Humanities, exceeds the current workforce capacity.

The MLLP workshop was designed to meet this demand by accomplishing the following outcomes for its undergraduate and graduate student attendees: students

·       Learned basic ML concepts through high-level talks from research experts at UArizona

·       Learned how ML is being applied and has advanced research in their domain

·       Gained experience in working collaboratively in inter- and multidisciplinary teams

·       Researched, prepared and delivered short presentations on their findings to a diverse audience

·       Learned about other opportunities at UArizona for data science learning and applications

There were over 80 participants from 25 majors at the event in addition to over 20 volunteers and 3 speakers.  Students learned about and presented on topics ranging from ML-assisted surgery to ML usage in Linguistics and Education.

Workshop schedule and information: https://sites.google.com/view/mlhackathontest/home

Data Science at UA: Open House

Tuesday, October 22, 2019

Updates from Data7, TRIPODS, and CyVerse

Learn about recent efforts by the UA Data Science Institute, Transdisciplinary Research in the Principles of Data Science (TRIPODS) Initiative, and the CyVerse project. Chat with faculty and staff involved in these groups and explore possibilities for enhancing your research through collaboration with them. Come to our Open House on Tuesday, November 5, 2019 from 3:00pm to 5:00pm in the Rincon Room of the Student Union.

The Latest Opportunities in Data Science

Thursday, August 15, 2019

Learn how a UA Data Science Ambassador can help your College with Data Science Literacy!

Read about Google's PhD Fellowship Program

First and Second Cohorts of Data Science Ambassadors

Monday, April 29, 2019

 

All Colleges were invited to nominate 2019-2020 Data Science Ambassadors. Deadline June 1, 2019.

2018-2019 Data Science Ambassadors: Champions for Data Science Literacy
--Kristy Makansi, RDI Communications

From astronomy to zoology, researchers use data to better understand their subject areas, develop and test hypotheses, and make new discoveries. But not every researcher is a data science expert. That’s where the Data Science Ambassadors come in.

Data Science Ambassadors (DSAs) are graduate students with the requisite knowledge and expertise to help researchers in their respective colleges develop data science skills. With a $1,000 stipend for an academic year commitment, DSAs receive training and support as they work to help others in their subject areas tap into the potential of data science.

Under the direction of Jeffrey Oliver and Vignesh Subbian, the DSA program is currently supported by and offered in the Colleges of Agriculture & Life Sciences, Engineering, Science, and Social & Behavioral Sciences.

Brian Maitner, a DSA in the College of Science, is a PhD student in Ecology & Evolutionary Biology who studies the processes that generate and maintain biodiversity at a global scale. Data science is critical to his field because “a single person can’t possibly collect the type of data needed to address the questions I’m interested in. Data science is needed to cobble together enough data in a useful way to address global questions.”

As an ambassador, Brian has provided coding help, particularly emphasizing Open Source solutions, and is running a weekly NetLogo workshop. Additionally, he’s working with the Introductory Biology Lab Director, Ryan Ruboyianes, to develop simulations that students can run on their own computers and smart phones and that help clarify some of the important ideas presented during lab.

A phenoclimatologist in the School of Natural Resources and Environment and the Laboratory of Tree-Ring Research, Amy Hudson is a DSA in the College of Agriculture and Life Sciences. As part of her DSA work, she and another ambassador, Jiali Han, have been instrumental in helping organize the first Women in Data Science – Tucson event. Hudson is grateful to the role other researchers have played in helping her learn to code and to “untangle an analysis.” Being an ambassador, she says, is a way to “pay these actions of mentorship forward in a more structured framework.”

Since becoming an ambassador, Hudson has been an instructor for two Data Software Carpentry workshops, including one titled Geospatial Lessons in R. When she arrived at the classroom on the first day, she saw that six of the 30 learners were fellow students in her department. She jumped in and approached the course as a cohort event. “We were laughing and groaning over learning this new skill together. Everyone was supporting each other, and it really made me proud and happy to be there to help out.”

For Joseph Long, whose work focuses on high-contrast astronomical imaging in the search for exoplanets, data science is one of the tools that makes his work possible.

“Data science is an umbrella term for a set of computational and statistical techniques, combined with the infrastructure for applying them to large data sets,” Long says when asked to describe data science. And Long certainly works with large data sets. His current project involves applying dimensionality reduction techniques to sequences of tens of thousands of images of a single star to model patterns of starlight and then subtract those patterns in the hopes of revealing a planet in the star’s orbit.

One of the ways Long is helping others leverage the power of data science in their own work is to get people out of their offices to share challenges, lessons learned, and insights gained.

“In addition to organizing a monthly talk/tutorial series on computing topics that is open to the whole university community, I set up a Slack channel to facilitate sharing questions and fostering discussion within our graduate student Slack organization.”

For Maitner, Hudson, and Long, participating in the DSA program is an effective way to spread the word about data science while also enhancing their understanding of the tools that enable their own work. Other DSAs include Jiali Han (College of Engineering), Xiang Liu (College of Agriculture & Life Sciences), Don Merson (College of Social and Behavioral Sciences), Matt Miller (College of Science), and Cristian Román-Palacios (College of Science).

If you would like to work with a DSA, submit a request at datascience@email.arizona.edu. If you would like your college to get involved in the DSA program, please contact one of the program directors for information.

Jeffrey Oliver, jcoliver@email.arizona.edu

Vignesh Subbian, vsubbian@email.arizona.edu

Data Science Fellow has a Hand in the First Image of a Black Hole

Friday, April 12, 2019

 

Data7 has a hat off to CK Chan, our Data Science Fellow who leads the Computations and Software Working Group for the Event Horizon Telescope project. This worldwide effort helps astronomers study objects predicted by Einstein's theory of General Relativity. You can read more HERE: https://eventhorizontelescope.org/

UA Center for Innovation in Brain Science awarded a $1.8 million NIH grant

Wednesday, April 10, 2019

 

Read the exciting news about the Center for Innovation in Brain Science!

Women in Data Science Tucson Regional Event

Tuesday, February 19, 2019

 

The Women in Data Science (WiDS) initiative aims to inspire and educate data scientists worldwide, regardless of gender, and support women in the field. WiDS started as a conference at Stanford in November 2015. Now, WiDS includes a global conference, with 150+ regional events worldwide; a datathon, encouraging participants to hone their skills; and a podcast, featuring leaders in the field talking about their work, and their journeys. On Friday, April 5th, 2019 WiDS-Tucson was held in the ENR2 building room S-107 from 8:30am-4:00pm, with a Happy Hour following.

WiDS-Tucson 2019 was amazing! Over 100 people attended and enjoyed a day of learning about Data Science efforts in many fields, with a lot of networking to share ideas. A very big Thank You to our Speakers, Data Blitz, and/or a poster presenters. All presenters did an excellent job and many positive comments were heard throughout the day. If you missed some of the sessions, here are links to recordings:

Aspinall Keynote, Kennedy, Zhang
Prof Dev Panel, AM DataBlitz Talks
PM DataBlitz, Borens Keynote, Hurwitz, Ida, Inspiration Panel
LaFleur, Riemer PM Concurrent Session

Look for our next WiDS-Tucson in the spring of 2020! To receive updates, subscribe to the WiDS-Tucson email list here: http://eepurl.com/gnSmOD 

You can Follow @TucsonWids on Twitter and "Like" our Facebook page: https://www.facebook.com/Wids-tucson-371018400385649/

If you're interested in joining the WiDS-Tucson planning committee, join our listserv: https://list.arizona.edu/sympa/info/wids-tucson/

Check out the carousel of photos from WiDS-Tucson 2019 below:

Pilot Project: Using NLP and ML to help Cancer Survivors

Tuesday, January 1, 2019

Can cancer survivors be motivated to adopt healthy behaviors? That has been the goal of the Lifestyle Intervention for Ovarian Cancer Enhanced Survival (LIVES) study, funded by NRG Oncology (GOG-0225 NCT 00719303)  and the National Cancer Institute (1R01CA186700-01A1 PI: Thomson) since 2012. This year the study reached a landmark in ovarian cancer research, enrolling their 1204th and final patient, making it the largest intervention study to date in ovarian cancer survivors. LIVES researchers are evaluating whether a low-fat diet high in vegetables, fruit and fiber with the addition of 4,000 physical steps per day over 24 months can reduce disease progression in women who have recently completed treatment for ovarian cancer compared to women who receive a general health education intervention.

 
Participants are paired with a telephone-based health coach who encourages them to exercise and to eat a healthy diet, through regularly scheduled phone conversations. The LIVES study has garnered over 10,000 hours of such exchanges. Can Data Science techniques help predict outcomes of these interventions based on sentiment analysis of these conversations? Can we find better ways to coach individuals that would increase likelihoods of successful behavior modification?  
 
photo of Dr. Tracy Crane
This project was among those selected for support in the form of a seed grant of $50,000 by the UA Data Science Institute (Data7), from white paper submissions received in early 2018. A Data7 team is working with Dr. Tracy Crane from the College of Nursing and Dr. Peter Jansen from the School of Information to dive into the data to see what can be uncovered, learned and improved upon in the science of behavior change.  
photo of Dr. Peter Jansen
The first challenge is to separate out the participant’s and coach’s spoken words from audio conversations. Once this has been achieved, Dr. Jansen will employ Natural Language Processing and Machine Learning techniques to glean useful insights from the data. When these data are paired with outcome data from the trial, including quality of life, diet and physical activity data, biomarker data, demographics and clinical data, the opportunities are boundless for exploring what it takes to change human behavior to improve health and ultimately prevent cancer.  

This project will provide foundational tools and techniques to process audio streams in near real time, providing assessments of conversations, with novel strategies that can be employed in other health intervention studies. Drs. Crane and Jansen will be utilizing outcomes of this endeavor to apply for an NIH Director’s New Innovator Award (DP2) in 2019.