QONTEXT · Quantum Contextual Information Access and Retrieval

The huge number and diversity of the users, the advertising products and services, the rapid growth of online resources have imposed new challenges to the conventional search. Queries are becoming even more broad and complex essentially due to context of the search process (e.g. the system, user, language, the word and action meaning, socio-psychological dimensions, interface and interaction methods). Over the past three decades, the research in IAR had led to various search engine models, such as vector space and probabilistic models. Unfortunately there has been no comprehensive investigation at the theoretical level for effectively integrating elements of context to create advanced search technology. The key issue preventing such research is a lack of a unified theoretical framework to seamlessly integrate the dimensions of context into the search engine models and into the evaluation protocols. This proposal is hinged on the belief that the dimensions of context can be naturally integrated into a generic and fundamental framework. To address the challenges of the dimensions of context in IAR this proposal shows a new vision of the IAR paradigm based on Quantum Theory (QT). This proposal starts from Van Rijsbergen’s seminal book. QT allows to measure relevance and context via projection, and probability of relevance via the trace, to logically reason through lattice of document structures and links, to change context via unitary operators, to handle correlations dependencies as density operators, to represent composite, entangled documents and features for which classical correlations cannot straightforwardly be used. The work independently done so far for some years by the partners suggest that more effective results can be obtained only if the expertises are exchanged through a network which would allow the partners to work together and exchange a wide range of expertises which is hardly possessed by a single team.