Healthcare monitoring system in hospitals and numerous health centers has experienced large growth, and portable healthcare monitoring systems with promising technologies are becoming of great concern. The huge volume of data generated by IoT devices creates a significant challenge in the healthcare system pertaining for handling and managing the data. The conventional techniques managing huge volume of data but accurate healthcare monitoring were not obtained since it learns the numerous features from raw inputs. In order to improve the healthcare monitoring system with big data, a novel technique called a TArgeted feature projection-based Hedge Iterated Gaussian Naive Bayes MIL Boost Classification (TAHIB) technique is introduced for patient e-health monitoring with lesser time. IoT patient data is gathered from the database. Initially in TAHIB technique, feature selection process is carried out using Motyka Indexive targeted projection method. The target projection method selects the relevant features through determining the similarity between features and objective (i.e., patient health monitoring). When similarity value is higher value, the feature is selected as relevant feature. After that, the relevant features are taken for patient e-health monitoring. Then, Hedge Iterated Gaussian Naive Bayes MIL Boost Classification process is carried out to classify the patient as normal or abnormal condition. The boosting classifier is an ensemble of several weak learners and combined to make a strong classifier hence it provides the final accurate classification results as normal or abnormal condition. Experimental evaluation of is carried out using on factors such as classification accuracy, classification time and error rate, space complexity with respect to number of patient data. The quantitatively discussed results indicate that performance of proposed TAHIB method increases data accuracy of disease diagnosis with a minimum time, error and memory consumption than conventional methods.