itc134fl17-waffle-pie

A project to create a webpage about Machine Learning, based on a longer whitepaper and research.
Group members: Pete Soukup, Tianyiru (Eve) Chen, Joseph Wanderer, Matthew Budd, Marianne Goldin
Method:The page was adapated for the web from a longer written document using PIE (point, illustration, example) outlining principles, and maybe waffles.
Read the page here: http://edison.seattlecentral.edu/~psoukup/itc134fl17-waffle-pie/index.html

Table of Contents

Machine Learning and its impact on jobs

1. Introduction
2. Examples
3. Process
4. Types
5. Effects on the Job Market
6. Further Thoughts

P.I.E.-Style Outline Example

Machine Learning

Introduction (Point: ML is incredibly powerful/useful)

• Point
  • Machine Learning is very broadly useful and core to AI because it allows programs to learn and adapt appropriately to new input without hardcoding its behavior
  • But to work, developers need to give it a lot of the right kind of data
• Illustration
  • Black Box Models: Machine Learning Algorithms discern patterns in data that would be difficult or impossible to code explicitly
• Explanation
  • ML models can then make accurate predictions, approriate decision or novel insights when confronted with new data, without anyone knowing how exactly.
  • ML makes large amounts of data much more useful and valuable to us

Examples (Illustration: Some of the many uses. ML pops up in many new disruptive technologies)

Examples graphic

• Point
  • Autonomous Vehicle technology is a very useful application of ML
  • ML is well suited to the problem of object recognition needed for autonomous vehicles
• Illustration
  • Object recognition for Autonomous Vehicles is very hard, must be accurate and must be appropriate to the task ie. vehicles, pedestrians, street signs etc.
• Explanation
  • With ML, object recognition models can be trained, tested and improved through trial and error by inputing more data and correcting mistakes.
  • Eventually, an accurate and robust object recognition model can be created without having to explicitly account for all possible circumstances/parameters on the road