A NOVEL METHOD TO EVALUATE CLUSTERING OF UNCERTAIN DATA USING KNN APPROACH
| Main Author: | N.Mangalam |
|---|---|
| Format: | Article |
| Terbitan: |
, 2019
|
| Subjects: | |
| Online Access: |
https://zenodo.org/record/2653592 |
Daftar Isi:
- Clustering is the process of making the group of abstract objects into classes of similar objects. A cluster of data objects can be treated as one group. Clustering of uncertain data has been well recognized as an important issue. This research paper proposes clustering of uncertain datasets, where similarities can be measured between datasets according to the characteristics. First, the user is registered with the server and the server verifies the user’s details with the database. After verification user sends the uncertain data to the server. The server uses KLL divergence mechanism for classifying discrete and continuous case data and computes the similarity of the data. Finally, apply K-NN algorithm to compute the distance between the nearest nodes and cluster the data. This method provides efficient clustering of uncertain data compared to other clustering methods. In the proposed system, KL (Kullback-Leibler) divergence and KNN approach are used together to overcome the existing drawback and to produce an effective performance with less time complexity.