New web information retrieval paradigm based on a multi-space interpretation index and projection operations

Classical search engines rely mainly on keyword matches. More advanced search engines and information retrieval systems use personalization techniques to enhance the relevancy of the results. Those approaches assume that the current search of a user is directly related to his/her profiles and/or past navigation/search experience. However, this assumption does not hold in all cases. In this paper, we are proposing an information retrieval approach that is not making any assumption about the user. It offers a straightforward way to choose and navigate the available interpretations. Our approach relies on a navigation model that is based on a new paradigm called multi-space interpretation index and projection operations. We have derived a mathematical model of a Web information retrieval system based on the new paradigm. This concept naturally distinguishes between the different interpretation layers (each layer is a separate space). Therefore, it offers to users the flexibility to navigate the interpretation spaces in such a way that suites better their information needs.