DMCA
A Weighted Nearest Neighbor Algorithm for Learning with Symbolic Features (1993)
| Venue: | Machine Learning |
| Citations: | 304 - 3 self |
BibTeX
@INPROCEEDINGS{Cost93aweighted,
author = {Scott Cost and Steven Salzberg},
title = {A Weighted Nearest Neighbor Algorithm for Learning with Symbolic Features},
booktitle = {Machine Learning},
year = {1993},
pages = {57--78}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
In the past, nearest neighbor algorithms for learning from examples have worked best in domains in which all features had numeric values. In such domains, the examples can be treated as points and distance metrics can use standard definitions. In symbolic domains, a more sophisticated treatment of the feature space is required. We introduce a nearest neighbor algorithm for learning in domains with symbolic features. Our algorithm calculates distance tables that allow it to produce real-valued distances between instances, and attaches weights to the instances to further modify the structure of feature space. We show that this technique produces excellent classification accuracy on three problems that have been studied by machine learning researchers: predicting protein secondary structure, identifying DNA promoter sequences, and pronouncing English text. Direct experimental comparisons with the other learning algorithms show that our nearest neighbor algorithm is comparable or superior ...
Keyphrases
symbolic feature weighted nearest neighbor algorithm feature space neighbor algorithm algorithm show protein secondary structure numeric value excellent classification accuracy sophisticated treatment direct experimental comparison attache weight english text real-valued distance nearest neighbor algorithm standard definition distance table distance metric symbolic domain dna promoter sequence
