DSS Blog

Recapping DSSVirtual with a Data ‘Bang’ in Healthcare, Finance & Technology

Written by Data Science Salon | Feb 26, 2021 7:49:17 PM

The very first DSSVirtual of 2021 was held February 16-17 - and we’re still feeling excited about how it turned out! With a diverse lineup featuring 20+ talks from leading data scientists, over 1,500 AI enthusiasts joining the virtual event, and lots of great networking, we can look back and say: it’s been a great one!

Here’s a recap, including some hot topics discussed, takeaways, and our favorite highlights.

Session highlights

For the first time ever, two tracks were available for attendees: the Technical Track, and the Business Track, both focused on data science applications in the healthcare, finance and technology industries. Attendees had the ability to choose between two sessions in one timeslot to custom tailor the event to their individual technical background and interests.

Coronavirus Hotspot Analysis

Ankur Chaudhary, Senior Data Scientist at Healthfirst, presented his research on Coronavirus hotspots and gave us a look at which regions, and who, in New York were at a higher risk of infection. This analysis included data that one might not expect, such as data on social vulnerability, social connectedness, comorbidities and underlying conditions, and more. Through this analysis, Ankur was able to provide us with actionable recommendations to keep the population safe, which continue to remain relevant with the current situation.

Addressing Disparities with Public Data on Social Determinants

Social factors in health was a topic also touched upon by other speakers at the event, such as Dax Paramanathan, Staff Data Scientist at Cityblock, who gave a talk about social determinants of health (SDOH). SDOHs are non-medical factors that affect overall health outcomes. One could imagine sunlight and fresh air being two examples, but this also ties into socioeconomic factors, financial factors, and more. This talk allowed us to understand how data can be used to connect an issue such as homelessness to overall health, and how we can integrate this kind of data into our machine learning models in order to best serve our communities, regardless of their situation, and to better fight inequality.

Transfer Learning: Applications in the Real World

We also learned that Transfer Learning is used to apply a trained model to a new (but related) problem. This turns out to be incredibly helpful when you might not have the time or dataset to train the model from scratch - such as in a hospital setting where data is still being gathered in a different department that might not be accessible. Thank you to Ayda Farhadi, an experienced data scientist with a demonstrated history of working in the hospital & health care industry, for sharing this great information.

These incredible innovations give hope that new technologies will help us face COVID-19 and other major challenges in the healthcare industry even more effectively this year.

Leveraging AI and machine learning in finance and technology industries

Other talks focused on the use of machine learning in the finance and technology industries, such as the one by Mingjie Zhao, Data Scientist at Wells Fargo. Mingjie provided tips on how to train and deploy a model to best handle imbalanced datasets. Another technical deep dive was presented by Saeed Shoaraee, Lead Data Scientist at S&P Global, who talked about applied machine learning for spreading financial statements. He provided solutions for financial analysts on how to avoid manual spreading of financial statements and discussed how ProSpread™ leverages computer vision and natural language processing in order to free up analysts and help them focus on their high value add operations.

Speed networking and amazing prizes

Watching our amazing speakers was not the only way to get a sense of learning and human connection. Although we weren’t able to mingle face-to-face near a physical snack table in a physical lobby, the Data Science Salon Slack workspace was given with the conference ticket so that all attendees could login and chat with each other, speakers, and sponsors.

In addition to this, we had our “SnackClub” speed networking sessions at the end of each day, which was an hour-long opportunity for guests to chat with video. With the simple click of a button, a new participant appeared on the screen for a fun, lighthearted and interactive 4 minute video-chat session. Topic cards are presented on the screen to help break the ice and keep conversation flowing - although, in our experience, an exciting dynamic in all of the conversations was nowhere near lacking. We’re impressed with the way the Slack channels were lighting up with questions, insights and of course some great humor! 

We couldn’t end the fun without some swag, so all attendees had the chance to win incredible giveaway prizes from our sponsors including a Sonos Move Bluetooth Speaker, machine learning e-books and Data Science Salon high-knee socks! Many thanks to our sponsors and community partners Algorithmia, Iguazio, Rapid Insight, Ubi Ops, Tiger Graph, John Snow Labs, NVIDIA, insideBIGDATA and Packt for sharing amazing prizes and resources with the community! The drumroll still continues as we’re about to announce the lucky winners. Good luck to all who entered!

Did you miss the event and want to catch up with 20+ data science presentations? All sessions are available on-demand and you can still register here to access them.

We’re looking forward to bringing the community back together at our upcoming DSSVirtual experience from May 4-5, which will focus on AI and machine learning applications in the media, advertising and entertainment industries. Make sure to snag your spot here, we’re looking forward to seeing you all there!