Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Data measures that characterise classification problems
Dissertation (MEng)--University of Pretoria, 2008.
Saved in:
| Other Authors: | |
|---|---|
| Format: | Thesis |
| Published: |
University of Pretoria
2013
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613707291328512 |
|---|---|
| access_status_str | Open Access |
| author2 | Barnard, E. |
| author_browse | Barnard, E. |
| author_facet | Barnard, E. |
| collection | Thesis |
| dc_rights_str_mv | © University of Pretoria 2008 E1080/ |
| description | Dissertation (MEng)--University of Pretoria, 2008. |
| format | Thesis |
| id | oai:repository.up.ac.za:2263/27624 |
| institution | University of Pretoria (South Africa) |
| last_indexed | 2026-06-10T12:40:25.453Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from UPSpace — University of Pretoria Institutional Repository |
| publishDate | 2013 |
| publishDateRange | 2013 |
| publishDateSort | 2013 |
| publisher | University of Pretoria |
| publisherStr | University of Pretoria |
| record_format | dspace |
| source_str | UPSpace — University of Pretoria Institutional Repository |
| spelling | oai:repository.up.ac.za:2263/27624 Data measures that characterise classification problems Barnard, E. cmvdwalt@gmail.com Van der Walt, Christiaan Maarten Classifier selection Data measures Data characteristics Artificial data Data analysis Classification Supervised learning Pattern recognition Meta-classification Classification prediction UCTD Dissertation (MEng)--University of Pretoria, 2008. We have a wide-range of classifiers today that are employed in numerous applications, from credit scoring to speech-processing, with great technical and commercial success. No classifier, however, exists that will outperform all other classifiers on all classification tasks, and the process of classifier selection is still mainly one of trial and error. The optimal classifier for a classification task is determined by the characteristics of the data set employed; understanding the relationship between data characteristics and the performance of classifiers is therefore crucial to the process of classifier selection. Empirical and theoretical approaches have been employed in the literature to define this relationship. None of these approaches have, however, been very successful in accurately predicting or explaining classifier performance on real-world data. We use theoretical properties of classifiers to identify data characteristics that influence classifier performance; these data properties guide us in the development of measures that describe the relationship between data characteristics and classifier performance. We employ these data measures on real-world and artificial data to construct a meta-classification system. We use theoretical properties of classifiers to identify data characteristics that influence classifier performance; these data properties guide us in the development of measures that describe the relationship between data characteristics and classifier performance. We employ these data measures on real-world and artificial data to construct a meta-classification system. The purpose of this meta-classifier is two-fold: (1) to predict the classification performance of real-world classification tasks, and (2) to explain these predictions in order to gain insight into the properties of real-world data. We show that these data measures can be employed successfully to predict the classification performance of real-world data sets; these predictions are accurate in some instances but there is still unpredictable behaviour in other instances. We illustrate that these data measures can give valuable insight into the properties and data structures of real-world data; these insights are extremely valuable for high-dimensional classification problems. Electrical, Electronic and Computer Engineering unrestricted 2013-09-07T11:52:19Z 2008-09-09 2013-09-07T11:52:19Z 2008-04-09 2008-09-09 2008-08-29 Dissertation a 2008 E1080/gm http://hdl.handle.net/2263/27624 http://upetd.up.ac.za/thesis/available/etd-08292008-162648/ © University of Pretoria 2008 E1080/ application/pdf University of Pretoria |
| spellingShingle | Classifier selection Data measures Data characteristics Artificial data Data analysis Classification Supervised learning Pattern recognition Meta-classification Classification prediction UCTD Data measures that characterise classification problems |
| title | Data measures that characterise classification problems |
| title_full | Data measures that characterise classification problems |
| title_fullStr | Data measures that characterise classification problems |
| title_full_unstemmed | Data measures that characterise classification problems |
| title_short | Data measures that characterise classification problems |
| title_sort | data measures that characterise classification problems |
| topic | Classifier selection Data measures Data characteristics Artificial data Data analysis Classification Supervised learning Pattern recognition Meta-classification Classification prediction UCTD |
| url | http://hdl.handle.net/2263/27624 http://upetd.up.ac.za/thesis/available/etd-08292008-162648/ |