Ad Litem Consulting, Inc.

In the following article, Microsoft’s Kevin Larson provides a great overview of word recognition. As our industry depends upon OCR and OCW, it never hurts to understand a little more about what happens behind the scenes.

What’s really interesting is how programmers look toward the field of psychology and then apply what they learn to the software. Humans do a great job of recognizing words. Software endeavors to take the same approach.

The Synopsis:

The Science of Word Recognition
or how I learned to stop worrying and love the bouma

by Kevin Larson, July 2004
Advanced Reading Technology, Microsoft Corporation

“[Larson] starts by describing three major categories of word recognition models: the word shape model, and serial and parallel models of letter recognition. [He] presents representative data that was used as evidence to support each model. After all the evidence has been presented, [Larson] will evaluate the models in terms of their ability to support the data. And finally…describe some recent developments in word recognition and a more detailed model that is currently popular among psychologists.”

The Link:

The Science of Word Recognition

Scientists create models to see all the potential outcomes of a given experiment. As example, how many paths will a rat take when going through a maze. So what would happen if you overlaid video of 1000 attempts?

Titled “The 1K Project II“, this video shows 1000 replays running on the same track in the PC stunt racer TrackMania. Not only is it beautifully shot and edited, it does illustrates quite beautifully how computer simulations can illustrate that which otherwise might defy simple explanation. The music is nice, too.