In a new working paper on the effects of assortment size on purchase probabilities, my coauthors (Xueming Luo & Zhijie Lin) and I try to take advantage of a food delivery app’s delivery radius constraints that lead to quasi-exogenous variations in kitchen assortment sizes. In its most simplistic form, I visualized the following thought experiment: … Continue reading Approximate border identification
Companion Git repo In studying the construction of movie trailers, one of the more interesting components is figuring out how many scene changes (cuts) are included in the 2m30s of the typical trailer. While ideally one would want to match scenes from a trailer to scenes from the actual movie, we do not have access … Continue reading Identifying scene changes in videos
Companion Git repo In my current research with Alex Chaudhry, we study the relationship between trailer design and opening weekend box-office sales. One of the things we’re interested in is the relationship between how much of the movie is revealed by the trailer and the type of movie that is being advertised. If you think … Continue reading Detecting Common Story Arcs in Movies
I was a recent victim of an internet fraud. I want to document it here so that I can share this with other consumers as well as the companies that are unknowingly involved. I think this also raises an interesting ethical debate for advertising platforms. The story I recently needed to change an existing United … Continue reading Google Ads Airline Fraud
EVENT ANNOUNCEMENT There will be a data science meetup on Weds. 3/6 at 6:30 pm. It will be located at CoWork Oasis. Dr. Gustavo Arriaga will be giving a talk/demo about identifying hotel renovations using machine learning. Topics covered include: NLP techniques (phrase modeling, topic modeling, etc.) Classification methods (Random forests, neural networks, etc.) Refreshments … Continue reading Identifying hotel renovations using machine learning
Imagine a world where there is no trust. You have traveled a long way, perhaps on the Silk Road, and encounter a vendor that sells drinking water. Everyone traveling along this road is headed to a far-off land from which they will not return. You know there are 2 sources water this vendor could have … Continue reading Do online reviews *improve* quality? A companion blog for my newest research.
For my analytics students, JupyterLab Beta version has just been released. Follow the install instructions to install and read the documentation to run through its features. I think, for students coming from Excel/ other self-contained data packages, this is a great addition to Jupyter Notebooks. JupyterLab is now essentially an IDE where all your documents, … Continue reading JupyterLab Beta Released!