Grace Hopper Annual Conference Schedule

SESSIONS LIST

Data Science in Diverse Industry Applications: Presentations

October 16 12:00 pm-1:00 pm
General Assembly Theatre C Level Three GRBCC
TRACK: Data Science
Presentation / Lightning Talk
Increasing Honesty in Airbnb Reviews
12:00 PM - 12:15 PM
LEVEL: Intermediate

We study the bias in online reviews using two experiments on Airbnb. In the first experiment, we induce more consumers to leave reviews by offering a coupon and find that those induced to review report more negative experiences. In our second experiment, we remove the possibility of reciprocation in reviews by changing the rules of the review system. We find that this change also decreases bias in reviews.

Data Pipelines for Music Recommendations at Spotify
12:15 PM - 12:30 PM
LEVEL: Intermediate

With a plethora of choices in today’s world, recommendation systems play a key role in enhancing user experience with personalized suggestions. In this talk, i will explain the recommender system architecture at Spotify by tracking the lifecycle from a user click to recommendations. We will also discuss how Spotify addresses the YOLO problem: "You Only Listen Once", before judging recommendations and the challenges arising from optimizing for user engagement.

Using Rule Ensembles to Predict Credit Risk
12:30 PM - 12:45 PM
LEVEL: Intermediate

A rule ensemble is a machine learning technique that combines the power of ensemble methods with the interpretability of decision trees. At Intuit, we have used rule ensemble methods to solve many predictive-modeling problems. In this talk I will describe what a rule ensemble is and why it is a good technique to predict risk, such as a small business’s creditworthiness.

How do you define user value? A Data Science Story
12:45 PM - 1:00 PM
LEVEL: Intermediate

In a world of seemingly unlimited data, simple and well-aligned metrics have become invaluable. In this presentation, I will walk you through the evolution of Dropbox's measure of user value. From a data science perspective, I will cover outlining the correct problem, designing and implementing a clustering and regression model, and validating results. Along the way, we will discuss what separates textbook data science from real world data science.