Overlapping clustering for textual data

Publikationstyp
Conference Paper
Date Issued
2018-02
Sprache
English
Author(s)
Khazaei, Atefeh  
Ghasemzadeh, Mohammad  
Gollmann, Dieter 
Institut
Sicherheit in verteilten Anwendungen E-15-H  
TORE-URI
http://hdl.handle.net/11420/11011
Start Page
115
End Page
118
Citation
7th International Conference on Software and Computer Applications, ICSCA 2018: 115-118
Contribution to Conference
7th International Conference on Software and Computer Applications, ICSCA 2018  
Publisher DOI
10.1145/3185089.3185113
Scopus ID
2-s2.0-85048466510
Publisher
ACM
ISBN
978-1-4503-5414-1
Texts have inherent overlapping, therefore for clustering textual data, the overlapping clustering algorithms are more appropriate. In this regard, a major challenge is that they are very slow in clustering big volumes of textual data. Among others, OKM and OSOM are two important overlapping clustering algorithms. In this study, we have implemented and compared the performance of these two algorithms. The experimental results of our study show that OKM clusters have better overlap sizes when these algorithms are used for clustering textual data. Since both of them require much time to complete, none of these two algorithms is suitable for clustering textual data. Therefore we mastermind a fast overlapping version of SOM which is suitable for this purpose.
Subjects
OKM
OSOM
Overlapping clustering algorithm
Textual data
DDC Class
004: Computer Sciences