Wireless Sensor Network (WSN) refers to a network of devices that can communicate the informa- tion gathered from a monitored field through wireless links. WSNs measures environmental conditions like temperature, sound, pollution levels, humidity, wind, etc. WSN’s data fault detection is a challenging problem due to presence of sensors in unpredictable areas. The data fault detection has to be precise and accurate, so that it can be used for weather prediction, disease prediction etc. In recent years, Machine Learning plays a vital role in accurate fault Detection. To make the detection process accurate it is mandatory to reduce the number of input features to the Support Vector Machine (SVM). Minimum Redundancy Maximum Relevance (MRMR) which is an efficient feature selection algorithm is proposed in this work. The feature set is used to train the Support Vector Machine (SVM) to detect data faults. The data fault detection accuracy is calculated which shows the accuracy of 92.9% for smaller test data and accuracy of 97.2% for larger test data.