There has been proliferation of Internet of Things (IoT) in almost all facets of life such as medical, agriculture, transportation, manufacturing to living homes to name some. IoT consists of sensors, actuators which are connected through LAN/MAN/WAN to detect and sense vital parameters. Due to widespread use of IoT, data produced by IoT is huge and raw in nature. This data at sometimes deposited on local servers and in most of the cases deposited on cloud for processing due to its large size. The raw data generated need to be abstracted, analyzed, visualized, optimally and correctly, to derive the best and useful knowledge and pattern in order to maximize profits, main- tain efficiency and accuracy, provide better services etc. as applicable for business intelligence in different domain.
Data produced by different IoT sources is complex, are of different types, commonly referred as big data. It has different types of attributes based on domain of application. Data produces historically may not have same attributes and values as produced today due to the change in external conditions and requirements. The learning methods used previously may become in-effective or produces sub-optimal results. There may be set new methods available which may give optimal results as compared to previous techniques.
Machine learning (ML) is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. Ensemble learning (EL) is a ML approach of combining mul- tiple algorithms/classifiers of same types or different types to classify or predict the values on unseen data.