Skip to content

Latest commit

 

History

History
34 lines (16 loc) · 1002 Bytes

File metadata and controls

34 lines (16 loc) · 1002 Bytes

Project Proposal

Question/need:

  • Can I predict the success of Kickstarter Campaign with classification models?

  • This project's goal is to help entrepreneurs/campaigners increase their chances of launching a successful campaign by narrowing down the criteria to focus with classification models.

Data Description:

  • Dataset: Kickstarter dataset from the website Web Robots

  • Each unit is an individual Kickstarter Campaign.

  • Features: backer_count, category, converted_pledge_amount, country, created_at, deadline, goal, launched_at, pledged, spotlight, staff_pick

  • Target: Success of the campaign

Tools:

  • Sklearn classification models - Logistic Regression, Decision Trees, Random Forest, KNN, etc.
  • pandas and numpy for data manipulation
  • SQL for storage
  • possibly using Flask for production if time allows

MVP Goal:

  • A baseline classification model that predicts the success of a Kickstart campaign