Breast cancer disease classification using fuzzy-ID3 algorithm with FUZZYDBD method: automatic fuzzy database definition
Breast cancer becomes the second major cause of death among women cancer patients worldwide. Based on research conducted in 2019, there are approximately 250,000 women across the United States diagnosed with invasive breast cancer each year. The prevention of breast cancer remains a challenge in the...
Saved in:
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
Peerj Inc.
2021
|
Subjects: | |
Online Access: | http://umpir.ump.edu.my/id/eprint/31462/1/index.html http://umpir.ump.edu.my/id/eprint/31462/ https://peerj.com/articles/cs-427/ https://doi.org/10.7717/peerj-cs.427 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.ump.umpir.31462 |
---|---|
record_format |
eprints |
spelling |
my.ump.umpir.314622021-05-28T11:15:12Z http://umpir.ump.edu.my/id/eprint/31462/ Breast cancer disease classification using fuzzy-ID3 algorithm with FUZZYDBD method: automatic fuzzy database definition Nur Farahaina, Idris Mohd Arfian, Ismail QA75 Electronic computers. Computer science Breast cancer becomes the second major cause of death among women cancer patients worldwide. Based on research conducted in 2019, there are approximately 250,000 women across the United States diagnosed with invasive breast cancer each year. The prevention of breast cancer remains a challenge in the current world as the growth of breast cancer cells is a multistep process that involves multiple cell types. Early diagnosis and detection of breast cancer are among the greatest approaches to preventing cancer from spreading and increasing the survival rate. For more accurate and fast detection of breast cancer disease, automatic diagnostic methods are applied to conduct the breast cancer diagnosis. This paper proposed the fuzzy-ID3 (FID3) algorithm, a fuzzy decision tree as the classification method in breast cancer detection. This study aims to resolve the limitation of an existing method, ID3 algorithm that unable to classify the continuous-valued data and increase the classification accuracy of the decision tree. FID3 algorithm combined the fuzzy system and decision tree techniques with ID3 algorithm as the decision tree learning. FUZZYDBD method, an automatic fuzzy database definition method, would be used to design the fuzzy database for fuzzification of data in the FID3 algorithm. It was used to generate a predefined fuzzy database before the generation of the fuzzy rule base. The fuzzified dataset was applied in FID3 algorithm, which is the fuzzy version of the ID3 algorithm. The inference system of FID3 algorithm is simple with direct extraction of rules from generated tree to determine the classes for the new input instances. This study also analysed the results using three breast cancer datasets: WBCD (Original), WDBC (Diagnostic) and Coimbra. Furthermore, the comparison of FID3 algorithm with the existing methods is conducted to verify the proposed method’s capability and performance. This study identified that the combination of FID3 algorithm with FUZZYDBD method is reliable, robust and managed to perform well in breast cancer classification. Peerj Inc. 2021-05-04 Article PeerReviewed pdf en http://umpir.ump.edu.my/id/eprint/31462/1/index.html Nur Farahaina, Idris and Mohd Arfian, Ismail (2021) Breast cancer disease classification using fuzzy-ID3 algorithm with FUZZYDBD method: automatic fuzzy database definition. PeerJ Computer Science. ISSN 2376-5992 https://peerj.com/articles/cs-427/ https://doi.org/10.7717/peerj-cs.427 |
institution |
Universiti Malaysia Pahang |
building |
UMP Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Malaysia Pahang |
content_source |
UMP Institutional Repository |
url_provider |
http://umpir.ump.edu.my/ |
language |
English |
topic |
QA75 Electronic computers. Computer science |
spellingShingle |
QA75 Electronic computers. Computer science Nur Farahaina, Idris Mohd Arfian, Ismail Breast cancer disease classification using fuzzy-ID3 algorithm with FUZZYDBD method: automatic fuzzy database definition |
description |
Breast cancer becomes the second major cause of death among women cancer patients worldwide. Based on research conducted in 2019, there are approximately 250,000 women across the United States diagnosed with invasive breast cancer each year. The prevention of breast cancer remains a challenge in the current world as the growth of breast cancer cells is a multistep process that involves multiple cell types. Early diagnosis and detection of breast cancer are among the greatest approaches to preventing cancer from spreading and increasing the survival rate. For more accurate and fast detection of breast cancer disease, automatic diagnostic methods are applied to conduct the breast cancer diagnosis. This paper proposed the fuzzy-ID3 (FID3) algorithm, a fuzzy decision tree as the classification method in breast cancer detection. This study aims to resolve the limitation of an existing method, ID3 algorithm that unable to classify the continuous-valued data and increase the classification accuracy of the decision tree. FID3 algorithm combined the fuzzy system and decision tree techniques with ID3 algorithm as the decision tree learning. FUZZYDBD method, an automatic fuzzy database definition method, would be used to design the fuzzy database for fuzzification of data in the FID3 algorithm. It was used to generate a predefined fuzzy database before the generation of the fuzzy rule base. The fuzzified dataset was applied in FID3 algorithm, which is the fuzzy version of the ID3 algorithm. The inference system of FID3 algorithm is simple with direct extraction of rules from generated tree to determine the classes for the new input instances. This study also analysed the results using three breast cancer datasets: WBCD (Original), WDBC (Diagnostic) and Coimbra. Furthermore, the comparison of FID3 algorithm with the existing methods is conducted to verify the proposed method’s capability and performance. This study identified that the combination of FID3 algorithm with FUZZYDBD method is reliable, robust and managed to perform well in breast cancer classification. |
format |
Article |
author |
Nur Farahaina, Idris Mohd Arfian, Ismail |
author_facet |
Nur Farahaina, Idris Mohd Arfian, Ismail |
author_sort |
Nur Farahaina, Idris |
title |
Breast cancer disease classification using fuzzy-ID3 algorithm with FUZZYDBD method: automatic fuzzy database definition |
title_short |
Breast cancer disease classification using fuzzy-ID3 algorithm with FUZZYDBD method: automatic fuzzy database definition |
title_full |
Breast cancer disease classification using fuzzy-ID3 algorithm with FUZZYDBD method: automatic fuzzy database definition |
title_fullStr |
Breast cancer disease classification using fuzzy-ID3 algorithm with FUZZYDBD method: automatic fuzzy database definition |
title_full_unstemmed |
Breast cancer disease classification using fuzzy-ID3 algorithm with FUZZYDBD method: automatic fuzzy database definition |
title_sort |
breast cancer disease classification using fuzzy-id3 algorithm with fuzzydbd method: automatic fuzzy database definition |
publisher |
Peerj Inc. |
publishDate |
2021 |
url |
http://umpir.ump.edu.my/id/eprint/31462/1/index.html http://umpir.ump.edu.my/id/eprint/31462/ https://peerj.com/articles/cs-427/ https://doi.org/10.7717/peerj-cs.427 |
_version_ |
1701163155349045248 |
score |
13.209306 |