DATA MINING BASED STREAM MINING APPROACH
A STREAM MINING BASED APPROACH FOR DYNAMIC ENVIRONMENT USING K-MEANS++ ALGORITHM
Versandfertig in 6-10 Tagen
29,99 €
inkl. MwSt.
PAYBACK Punkte
15 °P sammeln!
The Clustering is one of the most important technique in data mining. It aims partitioning the data into groups of similar objects. That is refered to as clusters. This research compares the StreamKM++ algorithm with the existing work, such as AP, IAPKM and IAPNA. The StreamKM++ algorithm is a new clustering algorithm from the data stream and itto constructs a good clustering of the stream, using a small amount of memory and time.Many researchers have done their work with static clustering algorithm, but in real time the data is dynamic in nature. Such as blogs, web pages, audio and video, etc...
The Clustering is one of the most important technique in data mining. It aims partitioning the data into groups of similar objects. That is refered to as clusters. This research compares the StreamKM++ algorithm with the existing work, such as AP, IAPKM and IAPNA. The StreamKM++ algorithm is a new clustering algorithm from the data stream and itto constructs a good clustering of the stream, using a small amount of memory and time.Many researchers have done their work with static clustering algorithm, but in real time the data is dynamic in nature. Such as blogs, web pages, audio and video, etc., Hence, the conventional static technique doesn't support in real time environment. In this work, the StreamKM++ algorithm is used which achieves high clustering performance over traditional AP, IAPKM and IAPNA. The experimental result shows StreamKM++ algorithm achieves the best result compared with existing work. It has increased the average accuracy rate and reduced the computational time, memory and number of iterations.
Andere Kunden interessierten sich für
