Real-time data clustering based on KK-Stream algorithm

In computer science, streaming algorithms are algorithms for processing data streams in which the input is presented as a sequence of items and can be examined in only a few passes, typically just one. These algorithms are designed to operate with limited memory, generally logarithmic in the size of the stream and/or in the maximum value in the stream, and may also have limited processing time per item. Stream data collecting algorithms similar to CluStream exertion derived from K-means. This kind of clustering algorithm is incapable of reaching the desired clusters and cannot manage the irregularities that may occur. In addition, they need to know the number of clusters. To find best way to solve this concern, this article, KK-Stream suggests a clustering outline so as to carry out data stream by its congestion-based loom. This algorithm, by means of an online constituent of every contribution data verification, generates a map with the intention of acts as an offline network collect system. The algorithm uses an offline method to dynamically change a data stream. via using the multipart associations among factors, the sum of statistics and cluster construction, our algorithm can be proficient and successfully generate clusters in real time. Furthermore, this method is used to detect and eliminate scattered clusters by using mapping of data stream densities to increase efficiency, speed, and accuracy in time. This method allows the clustering speed of the data stream to be made without reducing the clustering quality. trial results explain with the purpose of our algorithm has a superior worth and competence in decision groups used for real-time streaming data.

کلیدواژه ها:

Online data ، offline data ، data stream ، data clustering ، CluStream ، K-means ، KK-Stream.