Samaher Al-Janabi - Department of Information Networks, Faculty of Information Technology (IT), University of Babylon Babylon ۰۰۹۶۴, Iraq
Ahmed Patel - School of SOFTAM, Faculty of Information Science and Technology (FTSM), Universiti Kebangsaan Malaysia, ۴۳۶۰۰ UKM , Malaysia
Hayder Fatlawi - Information Technology Center,University of Kufa Najaf ۰۰۹۶۴, Iraq
Ibrahim AlShourbaji - Computer Network Department, Computer Science and Information System College, Jazan University Jazan ۸۲۸۲۲-۶۶۴۹, Saudi Arabia

With the arrival of big data and cloud computing asa computing concept, it is becoming ever more critical toefficiently choose the most optimum machine on which toexecute a program, for example in the healthcare environment.This process of choice is also complicated by the fact thatnumerous machines are available as virtual machines. Hence,predicting the most optimum choice of machine based on atarget application is a challenge. Prediction techniques consumelarge amount of computing resources when operating withmulti-dimensional data that can cause long delays compoundedby cross validation process in evaluating and choosing the mostoptimum prediction model. We propose a model of predictiontechniques to predict and classify some of the health datasets toretrieve useful knowledge to illustrate how a data miner canchoose a suitable machine especially in cloud environment withgood accuracy in a timely manner. Our results show that theexecution time has an inverse relation with the use of resourcesof a machine and the accuracy of prediction could be differentfrom one machine to another using the same predictingtechnique and dataset.

Computer architectures, Data Miner, Predicting techniques, Cloud computing, and Healthcare Datasets