A Patent Technique of Jaccard Discrete (J-DIS) Similarity Clustering Algorithm
Traditionally, the classification object yields homogeneous object to separate cluster. Few authors investigated clustering based on k-Means to distinguish intrusions based on the particular class. Mostly, k-Means algorithm finds out similarity between the object based on distance vector for smalles...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Journal |
| Language: | English |
| Published: |
Elsevier Publications
2014
|
| Subjects: | |
| Online Access: | http://ur.aeu.edu.my/526/1/A%20PatentTechnique%20of%20Jaccard%20DiscreteJ-DIS%20Similarity%20Clustering%20Algorithm.pdf http://ur.aeu.edu.my/526/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Asia e University |
| Language: | English |
| Summary: | Traditionally, the classification object yields homogeneous object to separate cluster. Few authors investigated clustering based on k-Means to distinguish intrusions based on the particular class. Mostly, k-Means algorithm finds out similarity between the object based on distance vector for smallest dataset. We proposed a new approach Jaccard Discrete (J-DIS) based approach which is combines with k-Means to find most similar measures over features attribute values in a larger dataset. Further, this paper is describing best suitable larger dataset taken from KDD CUP-99 dataset [1].Moreover, the J-DIS k-Means approach can be applied over clinical informatics and wireless clustering based routing protocols. |
|---|