BENCHMARKING the EFFECTIVENESS of CATEGORICAL RESPONSE VARIABLE MODELS and THEIR VISUALIZATIONS on WEATHER DATA

• Activity Number: 121 - SPEED: Data Expo
• Type: Contributed
• Date/Time: Monday, July 30, 2018 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Computing
• Author(s): Kristen Bystrom* and Zhi Yuh Ou Yang and Lei Chen
• Companies: and Simon Fraser University and Simon Fraser University
• Keywords: Supervised Learning; Rare Events; Data Challenge; Visualization; Neural Network; Machine Learning

Abstract

This is a digital poster supporting our analysis of the 2018 Data Expo Challenge. At this stage, much of the work remains to be completed. Our research team consists of three undergraduate statistics students at Simon Fraser University.

We will compare the precision and accuracy of several supervised learning models in general and in the case of rare events . We will also compare the effectiveness of common visualizations used for each model, and other trade offs.