Estimation of Sample for Data Mining of Alliance Policy Rajeswari K.1, Kiruthika R.1 1Assistant Professor, Department of Computer Science Engineering, Mahendra Engineering College (Autonomous), Namakkal, Tamilnadu, India Online published on 16 March, 2018. Abstract There is expanding enthusiasm for sharing the experience of items and services on the web stage, and online networking has opened a route for item and specialist co-ops to comprehend their customers’ needs and desires. In the paper investigates audits by cloud purchasers that reflect customer's encounters with cloud services. The audits of around 6, 000 cloud benefit clients were broke down utilizing conclusion examination to distinguish the disposition of each survey, and to decide if the feeling communicated was sure, negative, or impartial. The examination utilized two data mining devices such as RapidMiner and KNIME, the outcomes were thought about. to Create four expectation models in the investigation to calculate the reaction of client's reviews. The proposed method depends on four directed machine learning calculations: K-Nearest Neighbor (k-NN), Naïve Bayes (NB), Random Tree (RT), and Random Forest (RF). Based on Experimental evaluations, proposed algorithm improves accuracy 7.82%, and reduces Error Rate 7.82% of the proposed framework contrasted than previous classifiers. Top Keywords Cloud Services, Data Mining, Reviews, KNIME, RapidMiner, K-Nearest Neighbor (k-NN), Nave Bayes, Random Tree, and Random Forest. Top |