Conversion rates – you are (most likely) computing them wrong

How hard can it be to compute conversion rate? Take the total number of users that converted and divide them with the total number of users. Done. Except… it’s a lot more complicated when you have any sort of significant time lag.
Prelude – a story
Fresh out of school I joined Spotify as the first data analyst. One of my first projects was to understand conversion rates.

Read Full Story

Jesse or Celine? Text Classification on Before Sunrise Dialog: Part II

In the last article, we converted movie dialog into numbers using a bag-of-words approach. Here we go in another direction and use an embedding model to represent text in a dense manner. So instead of capturing stylistic tendencies of Jesse and Celine (via word choice in a bag-of-words model), we will try to capture the semantic differences between their words (via an embedding model).

Read Full Story