Subspace Clustering on High Dimensional Data
By: Bhargav M
.
Contributor(s): Reddy, P. Venkateswarlu.
Publisher: Tamil Nadu i-manager's 2016Edition: Vol.3(3), May-July.Description: 18-26p.Subject(s): Computer EngineeringOnline resources: Click here
In:
i-manager's journal on cloud computing (JCC)Summary: Due to the increase of large amount of real world, it is a difficult task for the organizations, companies, etc., to extract relevant data from large amounts of necessary and unnecessary data. Many researches will be going on from the few decades onwards. In datamining there is a concept called clustering which will be used for the smaller datasets, where it effectively makes relevant data into clusters. But the problem will arise on the larger datasets, where it will face a complexity for grouping relevant data into cluster. In this paper, analysis have been referred from many of the algorithms like subscale algorithm for finding the dense region from the dataset and DBscan algorithm for making a cluster as a result it takes dataset as an input and scans a complete dataset. The problem occurs on the time complexity and performance and also, it will follow a sequential flow of database scan. So, it takes time for relevant data values as a cluster in final result. In this analysis, it will allow the complete dataset scans at a time, processing data in parallel manner. So, the resultant data is in effective manner in a lesser time. For the improvement of previous algorithm a Map Red based DBscan for reducing time complexity and performance improvement has been used.
| Item type | Current location | Call number | Status | Date due | Barcode | Item holds |
|---|---|---|---|---|---|---|
Articles Abstract Database
|
School of Engineering & Technology Archieval Section | Not for loan | 2019901 |
Due to the increase of large amount of real world, it is a difficult task for the organizations, companies, etc., to extract relevant data from large amounts of necessary and unnecessary data. Many researches will be going on from the few decades onwards. In datamining there is a concept called clustering which will be used for the smaller datasets, where it effectively makes relevant data into clusters. But the problem will arise on the larger datasets, where it will face a complexity for grouping relevant data into cluster. In this paper, analysis have been referred from many of the algorithms like subscale algorithm for finding the dense region from the dataset and DBscan algorithm for making a cluster as a result it takes dataset as an input and scans a complete dataset. The problem occurs on the time complexity and performance and also, it will follow a sequential flow of database scan. So, it takes time for relevant data values as a cluster in final result. In this analysis, it will allow the complete dataset scans at a time, processing data in parallel manner. So, the resultant data is in effective manner in a lesser time. For the improvement of previous algorithm a Map Red based DBscan for reducing time complexity and performance improvement has been used.
Click on an image to view it in the image viewer
There are no comments for this item.