Get the latest news
Subscribe to receive updates from the School of Data Science.
Over 600 people registered for the annual event, however since the conference was virtual this year (due to COVID-19) over 800 unique viewers tuned into the conference, hosted on the school’s website. The School of Data Science captured viewers from the United States, United Kingdom, Morocco, Brazil, Qatar, Nigeria, India, Singapore, Italy and Denmark.
The day was full of women in different areas of data science, different stages in their careers, and different leadership levels, all doing outstanding work in their fields. Learn more about the conference below.
The live virtual conference began at 12:00pm with opening keynote speaker Christine Borgman on “Big Data, Little Data, or No Data? A Social Science Perspective on Data Science.”
Borgman is a Distinguished Research Professor in Information Studies at UCLA and the Director of the Center for Knowledge Infrastructures at UCLA.
In her keynote presentation, Borgman discussed how scientists collect, use, reuse, and often lose their data. She defined data as “representations of observations, objects, or other entities used as evidence of phenomena for the purposes of research or scholarship."
How should data be created and reused? How should scientists plan hardware and software that will last over multi-decade periods? How do we incentivize researchers to create data for the long term? These are the questions that motivate Borgman, as she presses into the challenges of information policy.
"Infrastructures are fragile,” Borgman stated in her presentation. “They break quickly, but the invisible infrastructure - data, metadata, provenance - is what keeps data alive and going."
Borgman drew in evidence from astronomy, environmental sciences, sensor networks, biomedicine, and other fields as she discussed data infrastructures and how so much of the focus of today is on big data, when the reality is there are many fields where some or little data is the norm.
She ended her presentation encouraging the audience to invest in the invisible infrastructures and make them sustainable so that researchers can reuse data in the long-term.
Panel: "From UVA to Industry, How to Hack into Your Career"
The day continued with a panel featuring four UVA alumni, Katherine Schinkel, Hope McIntyre, Melissa Phillips, and Caroline Lurillo.
Katherine Schinkel (MSDS ‘16) is a Data Scientist at Stitch Fix. Hope McIntyre (MSDS ‘16) is a Lead Data Scientist at Storyblocks. Melissa Phillips (MSDS ‘20) is a Data Scientist at CCRI. Caroline Lurillo, who moderated the panel, is a Senior Data Scientist at Microsoft.
The panel began with each woman describing their decision to pursue data science. They continued with how they ended up in their current jobs and what their day-to-day work looks like.
Participants submitted questions, and at the end, each panelist led a Zoom breakout room session, where audience members shared thoughts, concerns, and questions. What are the resources you most recommend? What is the biggest takeaway from the MSDS program? What skills do you think are the most important? The panelists answered these questions and their advice.
“Enjoy the journey,” Katherine Schinkel said when asked what she would say to someone going into data science. “This is a lifelong pursuit if you are in this field.”
Panel: "AI in the Public Sector -- Challenges and Future Opportunities"
At 2:30pm, The Deloitte AI Institute for Government sponsored a panel on artificial intelligence in the public sector.
The panel featured four women who work at Deloitte. Christina Canavan is the Managing Director for Deloitte Advisory. Vinita Fordham is a Specialist Leader for Deloitte Consulting. Sarah Milson is a Senior Manager for Deloitte Consulting. Madison Gallagher, who moderated the panel, is a Manager for Deloitte Consulting.
Canavan focuses on addressing data quality and integrity in her work at Deloitte, as she leads data analysis, data profiling, statistical sampling, and estimation techniques. Fordham specializes in artificial intelligence and technology innovation areas. Milsom has advised public sector clients on challenges, strategic planning, and change management for over 12 years. Gallagher has experience in several policy areas, including economic development, financial services policy, and technology implementation.
All four women have seen and experienced the growth of artificial intelligence and how it has impacted their clients and work.
How should the government be focusing on AI and AI strategy? Where should we apply AI? How do we make AI explainable?
The panelists explored these questions and more during this session.
“I would like to see more of an intentional focus on the journey from pilot to scale when it comes to AI,” Milsom said. “There is a lot of enthusiasm around AI right now, but there does not always seem to be a plan from experimenting from AI to applying it at the enterprise level.”
Another question that came up was, “How do we create an ethical framework for AI that allows it to be an ongoing conversation rather than a series of boxes we check at the beginning of a project?”
In answering this question, Fordham stated that data scientists should always be monitoring the efficacy of the AI models and algorithms they create and use and ensuring it is ethical throughout the process.
As these women shared their perspectives from their careers, they all emphasized the importance of ethical technology, as AI continues to grow.
An Introduction to Data Linking
Rachel House, a Senior Data Scientist at S&P Global Artificial Intelligence, gave a presentation at 3:30pm on data linking.
House defined data linking as “the task of identifying, matching, and merging records from multiple datasets that correspond to the same entities.”
She gave two primary reasons for the need for data linking. The first is that we are generating an enormous amount of data every minute of every day. The second is that we need to make sense of this data.
There are also challenges when it comes to data linking. House detailed that since data is often incomplete and computational complexity grows quadratically, there is a lack of training data, and privacy and confidentiality can be key concerns.
In this hour, House gave the audience a look into what data linking is, why it is necessary, where its challenges lie, and how to apply data linking now.
For the closing keynote, Vicki Schmanske gave a presentation titled, “Accept the Challenge and Shatter the Norms.”
Schmanske is the Intelligence Group President of Leidos, where she provides solutions to Intelligence Community agencies.
She walked the audience through her career and what she has learned along the way. Vicki encouraged the audience that the goal does not have to be how do I get to the top?
“I am a firm believer that careers are a jungle gym, not a ladder,” Schmanske said. “Sometimes when you have your feet on multiple rungs, you have more stability. Lateral steps can be so valuable."
As a woman in leadership in data science, she spoke to the importance of making room at the table for other women and encouraging each other.
"Support other women in the workplace,” she said. “To the men in the audience, we need your advocacy and support in shattering these norms."
When asked what the biggest piece of advice she would give to the audience, Schmanske said to be authentic and bring your whole self to work.
“Don't be afraid of the uncomfortable conversations,” Schmanske said “It is okay to fail. That is the way you are going to grow. Take those risks."
Learn more about our corporate sponsors and their dedication to WiDS here.
Subscribe to receive updates from the School of Data Science.