Back to articles
Articles
Volume: 31 | Article ID: art00002
Image
Visual Analytic Process to Familiarize the Average Person with Ways to Apply Machine Learning
  DOI :  10.2352/ISSN.2470-1173.2019.1.VDA-676  Published OnlineJanuary 2019
Abstract

The everyday consumer is inundated with applications powered by machine learning. But in an ordinary day, do we encounter situations and choices which could also benefit from machine learning for which there is no specific tool invented yet? We describe scenarios where people without any machine learning background could find it useful to define their own solution which uses machine learning. Although machine learning is becoming ubiquitous, the average person is unaware of the steps involved. This abstraction makes sense, in many situations, such as traffic predictions, it is not necessary for the driver to know what machine learning algorithm is running. However, we consider examples where knowing how to incorporate machine learning into a problem would assist in decision making. We propose a workflow with operations leading to a final application. There are several challenges here, namely, the average consumer is not expected to have a mathematical background, nor is expected to acquire any additional background. To achieve this new utility, we use a visual analytic pipeline which integrates machine learning and the person. We use the IEEE VAST 2018 Challenge as a case study in which the user steps through the workflow. Finally, we envision the resulting application.

Subject Areas :
Views 25
Downloads 7
 articleview.views 25
 articleview.downloads 7
  Cite this article 

Andrew Tran, Yamini Dasu, Anna Baynes, "Visual Analytic Process to Familiarize the Average Person with Ways to Apply Machine Learningin Proc. IS&T Int’l. Symp. on Electronic Imaging: Visualization and Data Analysis,  2019,  pp 676-1 - 676-7,  https://doi.org/10.2352/ISSN.2470-1173.2019.1.VDA-676

 Copy citation
  Copyright statement 
Copyright © Society for Imaging Science and Technology 2019
72010604
Electronic Imaging
2470-1173
Society for Imaging Science and Technology
7003 Kilworth Lane, Springfield, VA 22151 USA