Sistema de recomendación para la compra de inmuebles en la web

Abstract

In the context of this research, data collection on properties for auction is conducted using web scraping techniques on selected legitimate websites. This process allows for the extraction of relevant information that enhances the important attributes of each property. Additionally, each data point is correlated with information about points of interest in the locality to identify the added values associated with the properties. From this analysis, crucial attributes for collaborative filtering are determined, which are fundamental to the research. Subsequently, data exploration and cleaning are performed to ensure the information is homogeneous, yielding useful and efficient results. The final stage involves implementing a machine learning algorithm to partition samples according to specific parameters, with the aim of identifying options that best meet the buyers' needs, thereby providing the most suitable alternatives for the end user.