Abstract: The data is the most valuable thing in this modern world of Information Technology. As we can see the day to day the data is increasing as each and every people using the World Wide Web. This all system generated data or may be the personal or informative data will get generated in a huge amount of size. That data will get stored at the data centers or on cloud. But those will get stored on the Hard Disk Drives in data centers. So in some situation if the HDD got crashed then we will have lost our data. This work proposes to develop the failure prediction of Hard disk drive. We have chosen the accuracy and review measurements, generally important to the issue, and tried a few learning strategies, Adaboost, Naive Bayes, Logistic Regression and Voting. Our investigation shows that while we can't accomplish close to 100% forecast precision utilizing ML with the present information we have accessible for HDDs, we can improve our expectation exactness over the standard methodology Keywords: Machine learning, Adaboost, Naive Bayes, Voting, Logistic Regression
Hard disk drive failure detection with recurrence quantification
analysis
