QoS based Machine Learning Algorithms for Clustering of Cloud Workloads: A Review

Sukhpal Singh, Inderveer Chana
2015 International Journal of Cloud-Computing and Super-Computing  
Data mining, known as knowledge discovery is a computer-assisted, analytical process of digging through and analyzing large amount of data and extracting knowledge from the data. Data mining technologies has made many industries like marketing, sales, healthcare organization, financial institutions etc. quite successful. It has a lot of benefits in various fields. It helps in the quick analysis of data and has improved the quality of decision making process and is used to turn information into
more » ... n information into actionable knowledge. In this paper, data mining process and knowledge discovery process is discussed. Three well known data mining classifier algorithms namely ID3, J48 and Naive Bayes are discussed and their performance has been evaluated using different parameters to find the best algorithm. Further, Naive Bayes classifier algorithm is used for classification of workloads based on different Quality of Service (QoS) parameters. used for clustering of workloads based on different QoS parameters. The organization of rest of this paper is as follows: Sec. 2 presents state of the art of machine learning algorithms. Sec. 3 describes the Naive Bayes classifier algorithm based workloads clustering. Sec. 4 presents the conclusion and future scope of this research work. Figure 3. Precision-recall characteristics Precision Rate -It is the fraction of retrieved Information that is relevant to the search. Precision Rate = Recall Rate -Recall in information retrieval is the fraction of the Information that are relevant to the query that are successfully retrieved. Recall Rate = Error rate -The degree of errors encountered during extract information from a data set and transforms it into an understandable structure. In Table 1 , comparison is done with respect to time taken to build the model, along with the error rate for all the three data mining algorithms.
doi:10.21742/ijcs.2016.3.1.03 fatcat:odui3bqvpffunldagksbf6rbxa