The internet has become a significant transaction platform for the real estate industry. While it does urge users to search extensively for properties, it also wastes time, energy and may not provide relevant results. Recommendation system technology can be applied to ease the search. The following research presents the design and implementation of a house recommendation system, using a preference-based search that provides similar properties as recommendations. This research adopts a Content-based filtering approach using K-Nearest Neighbour, which is developed as a Web application with a Flask web framework using Python programming language. An end user demonstration shows the effectiveness of the system that uses machine learning as suppose to the system that does not use the same. It has been concluded that more than sixty per cent of the results are relevant to the user’s preferences. The limitations of the current research and implementations are discussed for future reference.