Wireless Sensor Networks consist of several low-cost,low-energy sensor nodes that sense data about their cor-responding environment and transfer it towards the desti-nation through a sink or coordinator node. As the WSNsare power-constrained, thus efficient mechanisms should beincorporated for achieving energy conservation to increasethe overall network lifetime by minimizing the network load.Clustering can be applied to divide the network into clusterswith each cluster having a cluster head where data collec-tion is performed. In many cases, the temporal or spatialdata collected by the sensor nodes within the same clustermight provide redundant values and thus increase the over-all size of data packets to be sent towards the Cluster-head.Thus, there is a need to aggregate the data at the Cluster-head such that redundancy can be removed and data can becompressed with less number of packets to be transmittedwithin the network. We provide a survey of various aggrega-tion models incorporated with the help of Machine Learningtechniques and also propose a Priority-based Data Aggre-gation (PbDA) scheme using machine learning in WSN.