Thursday, February 13

 

2:30–5:00 p.m.

WIDS @ Penn Tour & Workshops

Join us at Amy Gutmann Hall, Penn Engineering’s new hub for data science, for two engaging kick-off workshops as part of the Women in Data Science (WiDS) @ Penn Conference. 

One workshop, titled “Data to Discovery: Exploring AI with a Patent Case Study, ChatGPT, and Generative Models,” will be led by Linda Zhao and Xinyu Xie from the University of Pennsylvania.

The other, “Bridging Language and Vision: A Hands-On Introduction to Vision-Language Models,” will be led by Artemis Panagopoulou of the School of Engineering and Applied Science (Penn Engineering).

Friday, February 14

 

8:30–9:00 a.m.

Check In + Grab-and-Go Breakfast

9:00–9:15 a.m.

Welcoming Remarks

9:15–10:05 a.m.

Keynote Address

10:05–10:35 a.m.

The Future of Work: Is It Here Yet?

Headlines warn about the existential risks of artificial intelligence and the mental health implications of social media, about struggling supply chains and growing inflation, and Big Tech’s influence in both our news and personal data.

In this talk, we will discuss these issues and more, including what are some of the many of the futures of work, how it’s already here for some, and its implications for work today.

10:35–10:50 a.m.

Break

10:50–11:20 a.m.

Towards Targeted Therapeutics by Decoding Sex Differences Using Big Brain Data

The talk will delve into how data science has revolutionized the investigation into the intricate landscape of the human brain. It will focus on the fascinating dimension of sex differences, with the goal of unraveling the complexities of how male and female brains differ anatomically and functionally. We will navigate through research findings that highlight structural differences that contribute to diverse cognitive abilities and susceptibilities between the sexes.

The talk will highlight the transformative potential of multi-modal data integration for elucidating complex but meaningful insights about the brain. By harnessing the power of big data in clinical research, we can gain a comprehensive understanding of sex-related variations, paving the way for more precise and personalized medical interventions. It will also provide a way in which other therapeutically important factors like race and ethnicity can be investigated, leading to more equitable medicine.

11:20–11:35 a.m.

RHETORIC AND RECEPTION: Driving Factors of Engagement in Trump and Biden’s Tweets

At its core, American democracy values the voices of the people and encourages civic participation. While previously, politics and presidents felt removed from citizens’ lives, the internet has made the relationship between presidents and their constituents far closer. Now, many presidents frequently use X (formerly Twitter) to answer questions, define policies, and share ideas.

Regarding the events of January 6, many have speculated that presidential communication through X fueled this violence. Thus, we aim to investigate how Biden and Trump’s tweets from 2020 until the insurrection differ in rhetoric and change over time, and how these characteristics affect user support.

We focus on Twitter because it is one of the sole platforms where presidents write original content and frequently communicate with the online community. We retrieve tweets from Biden and Trump with information about the raw text, time published, number of likes, and number of tweets. To quantify support, we use likes, which primarily convey support. To identify characteristics, we perform sentiment analysis and mine for emotion type, intensity, positivity, negativity, toxicity, and violence through a variety of models such as BERT, VADER, Detoxify, and Grievance Dictionary. Ultimately, after conducting a comparative analysis, we decide to use OpenAI’s moderation endpoint to mine for similar category scores. By comparing Tweet characteristics and tracking them over time, we can see what types of tweets garner the most engagement.

Our research will improve our understanding of factors that drive engagement. While Biden and Trump will not run against each other in future elections, the highly charged political environment of 2020 is still present today. Based on conclusions from our research, future candidates can tailor their social media strategies to better connect with voters, and social media platforms can consciously regulate political content to create a healthy environment for democratic discourse.

11:35–11:50 a.m.

Transforming the Retail Experience with Generative AI: From Personalization to Product Innovation 

Join us for an engaging session that explores how Generative AI can revolutionize the retail seller and customer experience in the telecommunications industry. We’ll discuss the vast opportunities that Gen AI offers, from enhancing seller performance to driving product innovation.

Learn about the technical challenges and solutions for integrating AI in retail, and discover how personalized training and coaching for retail experts can be achieved using AI. Gain insights into key metrics to track for measuring success and effectiveness.

This session is perfect for anyone interested in the cutting-edge advancements that are shaping the way we do business!

11:50 a.m.–12:30 p.m.

Lunch

12:30 – 1:00 p.m.

Breakout Discussion & Lunch

1:05–1:45 p.m.

AI in Society Panel

This session will dive into the profound ways artificial intelligence is reshaping industries and society, and address AI’s influence across diverse sectors, with a particular focus on its ethical, social, and governance implications.

A highlight of the discussion will be an exploration of AI’s pivotal role in advancing the FinTech industry.

1:45–2:00 p.m.

UPenn Student Lightning Talk with Q&A

2:00–2:15 p.m.

Identifying Low Acuity Emergency Department Visits with a Machine Learning Approach

Current approaches to identifying low acuity Emergency Department (ED) visits rely on diagnosis code rule-based algorithms, which can be inconsistent and fail to account for the complexity of patient conditions. We demonstrate that machine learning–based predictive modeling, trained on demographic and clinical data, substantially outperforms traditional code–based methods.

This talk will present the development and validation of our models, discuss their advantages over existing strategies, and highlight the need for continued research to ensure both accuracy and fairness in predicting low acuity ED visits.

2:15–2:50 p.m.

Academic Talk with Q&A

2:50–3:00 p.m.

Closing Remarks