Improving class noise detection and classification performance: a new two-filter CNDC model
Class noise is an important issue in classification with a lot of potential consequences. It can decrease the overall accuracy and increase the complexity of the induced model. This study investigates ensemble filtering, removing and relabeling noisy instances issues and proposes a new two-filter mo...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
Elsevier B.V.
2020
|
| Subjects: |
| _version_ | 1796865223093649408 |
|---|---|
| author | Nematzadeh, Zahra Ibrahim, Roliana Selamat, Ali |
| author_facet | Nematzadeh, Zahra Ibrahim, Roliana Selamat, Ali |
| author_sort | Nematzadeh, Zahra |
| collection | ePrints |
| description | Class noise is an important issue in classification with a lot of potential consequences. It can decrease the overall accuracy and increase the complexity of the induced model. This study investigates ensemble filtering, removing and relabeling noisy instances issues and proposes a new two-filter model for Class Noise Detection and Classification (CNDC). The proposed two-filter CNDC model comprises two major parts, which are noise detection and noise classification. The noise detection part involves ensemble and distance filtering to overcome ensemble issues. In latter part, a Removing-Relabeling (REM-REL) technique is proposed to enhance overall performance of noise classification. To evaluate the performance of the proposed model, several experiments were conducted on six real data sets. The proposed REM-REL technique was found to be successful to classify noisy instances. The final results showed that the proposed model led to a significant performance improvement compared with ensemble filtering. |
| first_indexed | 2024-03-05T20:53:41Z |
| format | Article |
| id | utm.eprints-91416 |
| institution | Universiti Teknologi Malaysia - ePrints |
| last_indexed | 2024-03-05T20:53:41Z |
| publishDate | 2020 |
| publisher | Elsevier B.V. |
| record_format | dspace |
| spelling | utm.eprints-914162021-06-30T12:16:13Z http://eprints.utm.my/91416/ Improving class noise detection and classification performance: a new two-filter CNDC model Nematzadeh, Zahra Ibrahim, Roliana Selamat, Ali QA75 Electronic computers. Computer science Class noise is an important issue in classification with a lot of potential consequences. It can decrease the overall accuracy and increase the complexity of the induced model. This study investigates ensemble filtering, removing and relabeling noisy instances issues and proposes a new two-filter model for Class Noise Detection and Classification (CNDC). The proposed two-filter CNDC model comprises two major parts, which are noise detection and noise classification. The noise detection part involves ensemble and distance filtering to overcome ensemble issues. In latter part, a Removing-Relabeling (REM-REL) technique is proposed to enhance overall performance of noise classification. To evaluate the performance of the proposed model, several experiments were conducted on six real data sets. The proposed REM-REL technique was found to be successful to classify noisy instances. The final results showed that the proposed model led to a significant performance improvement compared with ensemble filtering. Elsevier B.V. 2020-09 Article PeerReviewed Nematzadeh, Zahra and Ibrahim, Roliana and Selamat, Ali (2020) Improving class noise detection and classification performance: a new two-filter CNDC model. Applied Soft Computing, 94 . ISSN 1568-4946 http://dx.doi.org/10.1016/j.asoc.2020.106428 DOI:10.1016/j.asoc.2020.106428 |
| spellingShingle | QA75 Electronic computers. Computer science Nematzadeh, Zahra Ibrahim, Roliana Selamat, Ali Improving class noise detection and classification performance: a new two-filter CNDC model |
| title | Improving class noise detection and classification performance: a new two-filter CNDC model |
| title_full | Improving class noise detection and classification performance: a new two-filter CNDC model |
| title_fullStr | Improving class noise detection and classification performance: a new two-filter CNDC model |
| title_full_unstemmed | Improving class noise detection and classification performance: a new two-filter CNDC model |
| title_short | Improving class noise detection and classification performance: a new two-filter CNDC model |
| title_sort | improving class noise detection and classification performance a new two filter cndc model |
| topic | QA75 Electronic computers. Computer science |
| work_keys_str_mv | AT nematzadehzahra improvingclassnoisedetectionandclassificationperformanceanewtwofiltercndcmodel AT ibrahimroliana improvingclassnoisedetectionandclassificationperformanceanewtwofiltercndcmodel AT selamatali improvingclassnoisedetectionandclassificationperformanceanewtwofiltercndcmodel |