Internet of things (IoT) is a system framework comprising of several layers such as device layer, fog layer, cloud layer, and each layer performs a specific task. By 2020, it is estimated that the total number of Internet-connected devices being used will be amounting to 50 billion. As these numbers grow and technologies become more mature, the IoT generates big data with a variety of multiple modalities and varying data quality. The fog layer is the intermediate layer, which aggregates the data from multiple IoT nodes within an area foot-print. Performing Analytics and Intelligent processing of this big data at the Fog end is the key to developing smart IoT applications. This lecture emphasises upon the various machine-learning methods that deal with the challenges presented by IoT data and its efficient implementation in the IoT-Fog Computing framework by considering some of the emerging domains of IoT applications such as health, environment, agriculture etc.