SOM Classification Method based on Transduction Scheme

Author

Tong, Bin ; Qin, Zhi-guang ; Ma, Xin-xin ; Wang, Yong ; Jia, Wei-feng

Author_Institution

Sch. of Comput. Sci. & Eng., Univ. of Electron. Sci. & Technol. of China, Chengdu

fYear

2008

fDate

13-15 Dec. 2008

Firstpage

12

Lastpage

15

Abstract

Transductive confidence machines (TCMs) when used in classification problems can provide us with reliability for every classification. Many machine learning algorithms, such as KNN algorithm, etc., have been incorporated with TCM, while there´s no SOM classification method based on TCM. Considering properties of SOM map unit, this paper first designs a novel nonconformity measurement and TCM-SOM classification method; and then its classification accuracy that is much more better than that of SOM and is close or even higher than that of TCM-KNN is also proved by UCI machine learning datasets.

Keywords

learning (artificial intelligence); pattern classification; self-organising feature maps; KNN algorithm; SOM classification method; UCI machine learning dataset; machine learning algorithm; transductive confidence machine; Algorithm design and analysis; Computer science; Design methodology; Electronic mail; Error analysis; Machine learning; Machine learning algorithms; Reliability engineering; Support vector machine classification; Support vector machines; Machine learning; TCM-KNN; TCM-SOM; classification;

fLanguage

English

Publisher

ieee

Conference_Titel

Apperceiving Computing and Intelligence Analysis, 2008. ICACIA 2008. International Conference on

Conference_Location

Chengdu

Print_ISBN

978-1-4244-3427-5

Electronic_ISBN

978-1-4244-3426-8

Type

conf

DOI

10.1109/ICACIA.2008.4769960

Filename

4769960