Accessibility navigation


Identifying problematic classes in text classification

Roberts, P., Howroyd, J., Mitchell, R. and Ruiz, V. (2010) Identifying problematic classes in text classification. In: 9th IEEE International Conference on Cybernetic Intelligent Systems, 1-2 Sept 2010, Reading, UK, pp. 136-141.

Full text not archived in this repository.

It is advisable to refer to the publisher's version if you intend to cite from this work. See Guidance on citing.

Official URL: http://dx.doi.org/10.1109/UKRICIS.2010.5898142

Abstract/Summary

Real-world text classification tasks often suffer from poor class structure with many overlapping classes and blurred boundaries. Training data pooled from multiple sources tend to be inconsistent and contain erroneous labelling, leading to poor performance of standard text classifiers. The classification of health service products to specialized procurement classes is used to examine and quantify the extent of these problems. A novel method is presented to analyze the labelled data by selectively merging classes where there is not enough information for the classifier to distinguish them. Initial results show the method can identify the most problematic classes, which can be used either as a focus to improve the training data or to merge classes to increase confidence in the predicted results of the classifier.

Item Type:Conference or Workshop Item (Paper)
Refereed:Yes
Divisions:Science > School of Mathematical, Physical and Computational Sciences > Department of Computer Science
ID Code:17362
Additional Information:IEEE Conference ID Number: 17717

University Staff: Request a correction | Centaur Editors: Update this record

Page navigation