eVEREst is a project devoted to the subject of real estates value estimation. The aim of the project is to support the system storing data about real estates (e.g. cadastral system) with information filtered from external sources (i.e. Internet). This additional information would be used in the process of estimating the market value of real estates. The presumption of the system is that the value of real estate depends on several factors that can be divided into three groups: socio-economical (f.e. level of pollution, noise, crime, neighbourhood, income per capita in the area, type of buildings), infrastructural (motorways, media, communication) and environmental (natural disasters, parks, forests, landscape). Data and information about these factors can be found in structured (databases) and unstructured sources (press releases, web pages of news agencies, etc.). Based on this assumption eKW concept will be utilized. Documents filtered from the Internet will be automatically indexed in three different ways (cognitive, time and spatial) in order to enable bounding with specific object in the real estates database. In next step enhanced report about each of the real estates can be presented. In the second phase of the project an expert system that will enable automated value estimation is to be implemented.. eVEREst is based on technologies such as information retrieval, collaborative filtering, data warehouses. Moreover ontology and web services are introduced in order to increase interoperability among its parts.