AutoGPT Vulnerabilities
This blog page is where I am testing out AutoGPT’s vulnerabilities, as a demonstration of why it is unsafe to hook it up to anything important.
Posts by Year
2023
2022
Topic Modeling MTG Decklists
Earlier this year, the team at EDHREC.com was generous enough to share with me a snapshot of their data, in the form of a .csv file of over 400,000 decklists...
SQL Practice With D&D Data
Below are some SQL challenges I made for myself as practice, using a database I built from public D&D Beyond character sheets.
D&D Race/Class Combos in ~1 Million Character Sheets
Last year, I scraped data from more than 1 million D&D character sheets from online sources, storing them in a SQLite database. The project was both for ...
2021
‘Journals Like You Also Reviewed…’
By now, most people probably have an intuitive understanding of how a simple recommendation system works. Classical recommendation systems use what is known ...
Dash App for a Genre Classifier
About two years ago, I designed a simple scikit-learn genre classifier for the Textual Optics Lab. You can read about it at this blog post.
Authorless Topic Models
For the first chapter of my dissertation (and, hopefully soon, a published article) I’ve been working with a topic model trained on a corpus of 2,348 New Yor...
A Hierarchical Cluster Analysis of the Book Review Index
I recently produced a hierarchical cluster analysis of the Book Review Index data that I’ve been working with, though unfortunately I didn’t have enough time...
ACH 2021: Computational Analysis of Book Reviews
The following is a lightly-edited transcription of my presentation with Dan Sinykin for ACH 2021, part of a roundtable on “Computational Analysis of Book Rev...
2020
Genre Classification with Sklearn Permalink
I wrote about using Sklearn to classify novels by genre for the Textual Optics Lab’s research blog.