Skip to content

Code from Predictive Challenge #1 during December 17, 2015 kickoff event

Notifications You must be signed in to change notification settings

TechTank-DataScience/PredictiveModelsChallenge

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 

Repository files navigation

#Predictive Models Challenge

On December 17, 2015, the Tech Tank Data Science Cohort met to kickoff their class. What is Tech Tank? The Tech Tank Data Science Program Playbook says it very well:

The Tech Tank is designed to rapidly build a pool of high-demand data scientists in an innovative and differentiated way. The Program will produce:

  • A highly networked and visible cadre of high potential junior data science talent
  • Emerging leaders whose functional expertise is complemented with consulting skills
  • Professionals who can be deployed against clients’ most complex problems

This kickoff event included a Predictive Models Challenge. The cohort formed into teams of 4-5 people. Using a predefined challenge and data sets, each team was asked to plot the distributions of each possible outcome to create a predictive model and present results to their colleagues. The challenge had to objectives:

  • Demonstrate ability to synthesize disparate data to develop a predictability model
  • Demonstrate ability to collaborate successfully in a team to accomplish objectives in a timely manner

Teams were offered two different challenges to choose from. This repo contains the code for some of the solutions developed by the teams. Here is a description of both challanges:

##Titanic Challenge

  • Create a simple model for predicting the survivability of passengers on the historic Titanic disaster
  • Plot the distributions of each of the possible factors that could influence survival/non survival
  • Identify thresholds in the distributions to build a model that predicts survival/non survival

Additional details and datasets - https://www.kaggle.com/c/titanic

##Random Acts of Pizza

  • Create a simple model for predicting whether or not a person will receive a free pizza based on the text of their pizza request on the raopredditmessage boards
  • Plot the distributions of each of the possible factors that could influence the gift of pizza
  • Identify thresholds in the distributions to build a model that predicts pizza or no pizza

Additional details and datasets - https://www.kaggle.com/c/random-acts-of-pizza

About

Code from Predictive Challenge #1 during December 17, 2015 kickoff event

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published