This repository has been archived by the owner. It is now read-only.
No description or website provided.
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

CDS Kaggle Subteam - Fall 2017


Zillow is an online real estate company that revolutionized the real estate industry, by tracking large volumes of digital real estate information. They develop a set of Zestimates for properties across the United States, which they continually improve as they operate their main real estate operations (their online advertising platform). In this competition, Zillow has released an unprecedented amount of real estate data available to the general public. The objective for this competition is to develop a model that correctly estimates the log error of their own Zestimates - in other words, to predict how wrong Zestimates are.

Members: Mitchel Fung, Sena Katako, Max Jiang

The safe driver competition is designed by a Brazilian auto company, Porto Seguro, and the goal is to predict whether a driver will file an auto insurance claim in the next year. Evaluation in this competition is done based on the Normalized Gini Coefficient which ranges from 0, a result of random guessing to 1, perfect predictions. Ideally, at the end of this competition, Porto Seguro will be able to make more accurate auto predictions and this will hopefully increase insurance coverage to other drivers in Brazil.

Members: Kevin Luo, Lily Liu, Sean Hu, Aaron Lou

The goal for this Kaggle challenge is to build an algorithm that automatically identifies if a target found by remote sensing systems is a ship or iceberg. We need to analyze the shape, size and brightness of the object and its surrounding using data from two channels: HH (transmit/ receive horizontally) and HV (transmit horizontally/ receive vertically).

Members: James Chen, Joseph Chuang

Corporacion Favorita is an Ecuadorian supermarket chain - one of the three biggest chains in the country. They are faced with the task of stocking the correct amount of both perishable and nonperishable goods - understocking can lead to poor service, whereas overstocking can lead to wasted resources. Our challenge is to predict the sales quantities for various combinations of items, dates and stores, therefore helping the company predict how to stock their brick and mortar locations.