Abstract | ||
---|---|---|
For many real world problems we must perform classification under widely varying amounts of computational resources. For example, if asked to classify an instance taken from a bursty stream, we may have anywhere from several milliseconds to several minutes to return a class prediction. For such problems an anytime algorithm may be especially useful. In this work we show how we convert the ubiquitous nearest neighbor classifier into an anytime algorithm that can produce an instant classification, or if given the luxury of additional time, can continue computations to increase classification accuracy. We demonstrate the utility of our approach with a comprehensive set of experiments on data from diverse domains. We further show the utility of our work with two deployed applications, in classifying and counting fish, and in classifying insects. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1007/s10044-007-0098-2 | Pattern Anal. Appl. |
Keywords | DocType | Volume |
bursty stream,classifying insect,computational resource,comprehensive set,instant classification,non-parametric distance-based classification,additional time,class prediction,classification accuracy,diverse domain,real world problem,anytime classification � nearest neighbor applications,nearest neighbor | Journal | 11 |
Issue | ISSN | Citations |
3-4 | 1433-755X | 2 |
PageRank | References | Authors |
0.44 | 21 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Xiaopeng Xi | 1 | 531 | 24.76 |
Ken Ueno | 2 | 124 | 13.27 |
Eamonn J. Keogh | 3 | 11859 | 645.93 |
Dah-Jye Lee | 4 | 422 | 42.05 |