Multi-space Projection Based Search Engine: Theoretical Model Instantiation and Prototype

Information Retrieval (IR) is still one of the most important and active research areas in computer science. The most popular applications of IR are Search Engines that help users locate information in the Web. It has to be noted that they have some limitations such as results´ relevance and exploration. We propose a Multi-Space Search Engine model that is based on multiple interpretation spaces. To evaluate the effectiveness of the proposed model, we developed a prototype and conducted a comparative study that includes content interpretation, search features and usability. The experiments showed that our prototype performs well when compared to popular search engines with regard to the visibility of interpretations and the navigability of the results.