An Adaptive Learning Solution to the Keyboard Optimization Problem

Abstract

The authors consider the problem of assigning more than one symbol of a finite alphabet A to the same key on a keyboard. Since multiple symbols of the alphabet A reside on the same key, the representations of all the words in a finite dictionary H need not be unique. The problem is one of optimally assigning the symbols of the alphabet to the keys of a given keyboard with a view to minimize the total number of words that have ambiguous representation. The problem is proven to be NP-hard. After presenting the only reported solution to the problem, a fast learning-automaton-based solution to this problem is reported. Experimental results demonstrating the power of this solution are presented.$<>$

BibTeX

@article{journalArticle:oommenAdaptiveLearningSolution1991,
 abstract = {The authors consider the problem of assigning more than one symbol of a finite alphabet A to the same key on a keyboard. Since multiple symbols of the alphabet A reside on the same key, the representations of all the words in a finite dictionary H need not be unique. The problem is one of optimally assigning the symbols of the alphabet to the keys of a given keyboard with a view to minimize the total number of words that have ambiguous representation. The problem is proven to be NP-hard. After presenting the only reported solution to the problem, a fast learning-automaton-based solution to this problem is reported. Experimental results demonstrating the power of this solution are presented.$<>$},
 author = {Oommen, B. J. and Valiveti, R. S. and Zgierski, J. R.},
 date = {1991-11},
 doi = {10.1109/21.135704},
 issn = {0018-9472},
 journaltitle = {IEEE Transactions on Systems, Man, and Cybernetics},
 number = {6},
 pages = {1608--1618},
 title = {An Adaptive Learning Solution to the Keyboard Optimization Problem},
 volume = {21}
}