Selecting Green Data Mining Services

Nowadays, there is a big increase in the usage of data analytic applications and services because of the growth in the data produced from many different sources. The QoS properties such as response time, reliability and latency of these services are important factors to decide which services to select. As we know, the energy consumption is becoming a big issue as a result of IT expansion. Therefore, establishing a QoS-based web service selection approach that considers energy consumption as one of the essential QoS properties represents a significant step towards selecting the greener web service. This paper presents an experimental study of energy consumption and latency behavior of data mining algorithms running as web services. Our study shows that, there is a strong relation between the dataset properties such as dataset size, number of attributes, data type, and QoS attributes energy consumption and latency. Based on the findings from our study, a prediction system is built which can be used to predict the energy consumption and latency values for data mining web services on a given dataset, and then these services can be ranked according to their predicted energy and latency values. Experimental results show the effectiveness of our prediction and service selection system.