Abstract: Market for housing contributes significantly to a country economy. As the market grows and become important, it attracts individuals as well as investors to get involve with the market. For individuals and investors, decision in buying a house is quite a complex task. There are many aspects that require careful consideration due to capital intensive nature of housing market. One of the aspects that is considered by homeowners is the neighbourhood quality and this subsequently link to the concept of housing market segmentation or housing submarket. Prior research have shown that through hedonic model, housing is priced by its bundle of characteristic that includes physical, locational, environment, legal and many more. This approach of house price modelling considers market as an aggregate. However, aggregate modelling may not be able to capture the heterogeneity of the housing characteristics due to the house price fundamentally varied spatially. Prior literatures suggest that the aggregate model might lead to serious inaccuracy in the house price estimation and thus, a disaggregated model should be considered. In order to form a disaggregated model, large area of housing market should be segmented into several distinct submarkets. This study is to evaluate prior research on housing market segmentation. There is inconsistency in current research in defining the housing market segmentation. Two main approaches; a priori defined approach and statistical approach are widely being employed in defining the submarkets. The inclusion of submarkets are found to be significant in most study related to housing market segmentation as it helps in improving the house price prediction accuracy.