Attended Events - Data Science Guides
Welcome to Attended Events' Data Science documentation. Attended Events are gatherings with a start and end date and time, where people come together in one location for entertainment or business, take the 2020 Super Bowl game or San Diego Comic-Con as an example. PredictHQ uses machine-learning models to predict attendance and ranks for each of these events. This content is provided as a resource for Data Science teams to help get them up and running quickly.
We provide guides to using our API and data with common Data Science tools and libraries in Python. The articles include a link to Jupyter Notebooks that you can download and run. The guides include code samples and instructions for performing common tasks.
Attended Events are scheduled to occur at a specific venue and usually depend on attendance, such as conferences, expos, concerts, festivals, performing-arts, sports and community.
This How to Series consist of the following three notebooks, allows you to quickly extract the data (Part 1), explore the data (Part 2), and experiment with different aggregations (Part 3):
Part 1: Data Engineering shows you how to call the Events API to extract the data to a DataFrame.
Part 2: Data Exploration is a guide to exploring PredictHQ's Attended Events data for Data Science teams.
Part 3: Feature Engineering provides examples of how to aggregate and extract features from the data.
Handling multi-day and Umbrella events - information on how to handle multi-day events in your forecast
Feature engineering guide - use the features API to create ML features for attended events