Skip to content

Instantly share code, notes, and snippets.

anonymous
anonymous / startups.txt
Created Nov 16, 2017
startups likeliness to fail or succeed based on ml model
View startups.txt
This file has been truncated, but you can view the full file.
For this classifier, success was defined as achieving an acquisition, acqui-hire, or an IPO, and failure as either shutting down, or being founded before 2004 without experiencing an exit. Things fed to the classifier were startup name, categories (enterprise software, SaaS, Analytics, Advertising, etc), location, year founded, and VCs providing funding (and the rounds in which they participated).
The overall accuracy (AUC score) was 0.757015.
Succeed:
94.9043% dropbox
93.9504% airbnb
93.6332% yext
You can’t perform that action at this time.