n Information Retrieval (IR), whether implicitly or explicitly, queries and documents are often represented as vectors. However, it may be more beneficial to consider documents and/or queries as multidimensional objects. Our belief is this would allow building "truly" interactive IR systems, i.e., where interaction is fully incorporated in the IR framework. The probabilistic formalism of quantum physics represents events and densities as multidimensional objects. This paper presents our first step towards building an interactive IR framework upon this formalism, by stating how the first interaction of the retrieval process, when the user types a query, can be formalised.
n Information Retrieval (IR), whether implicitly or explicitly, queries and documents are often represented as vectors. However, it may be more beneficial to consider documents and/or queries as multidimensional objects. Our belief is this would allow building "truly" interactive IR systems, i.e., where interaction is fully incorporated in the IR framework. The probabilistic formalism of quantum physics represents events and densities as multidimensional objects. This paper presents our first step towards building an interactive IR framework upon this formalism, by stating how the first interaction of the retrieval process, when the user types a query, can be formalised.
Abstract The formalism of quantum physics is said to provide a sound basis for building a principled information retrieval framework. Such a framework is based on the notion of information need vector spaces, where events, such as document relevance, correspond to subspaces, and user information needs are represented as weighted sets of vectors. In this paper, we look at possible ways to build, using an algebra defined over weighted sets, sophisticated query representations and report on corresponding experiments on TREC. 1
The probabilistic formalism of quantum physics is said to provide a sound basis for building a principled information retrieval framework. Such a framework can be based on the notion of information need vector spaces where events, such as document relevance or observed user interactions, correspond to subspaces. As in quantum theory, a probability distribution over these subspaces is defined through weighted sets of state vectors (density operators), and used to represent the current view of the retrieval system on the user information need. Tensor spaces can be used to capture different aspects of information needs. Our evaluation shows that the framework can lead to acceptable performance in an ad-hoc retrieval task. Going beyond this, we discuss the potential of the framework for three active challenges in information retrieval, namely, interaction, novelty and diversity.
Even the best information retrieval model cannot always identify the most useful answers to a user query. This is in particular the case with web search systems, where it is known that users tend to minimise their effort to access relevant information. It is, however, believed that the interaction between users and a retrieval system, such as a web search engine, can be exploited to provide better answers to users. Interactive Information Retrieval (IR) systems, in which users access information through a series of interactions with the search system, are concerned with building models for IR, where interaction plays a central role. There are many possible interactions between a user and a search system, ranging from query (re)formulation to relevance feedback. However, capturing them within a single framework is difficult and previously proposed approaches have mostly focused on relevance feedback. In this paper, we propose a general framework for interactive IR that is able to capture the full interaction process in a principled way. Our approach relies upon a generalisation of the probability framework of quantum physics, whose strong geometric component can be a key towards a successful interactive IR model.
This thesis is fundamentally about using conceptual and mathematical constructs from the area of Quantum Theory in Information Retrieval (IR). The need and motivation for this is two-fold – firstly, it has been increasingly shown in decision sciences that human decision-making does not always conform to the norms of traditional probability and logic framework. The quantum framework offers a generalised probability and logic framework which can model decisions or judgements under dynamic context and ambiguity. Secondly, there is a need in IR for theories and models which improve our understanding of user behaviour. Hence it is worth exploring the combination of the quantum framework and IR, especially focused on the user aspects of IR. The overarching research question is whether there is evidence of user behaviour in IR scenarios which warrants the need for a quantum based approach by way of showing the limitations of the traditional (classical) approach. The methodology involves analysing data to detect quantum-like phenomena of interference, contextuality, incompatibility, etc. from two common data sources in IR–standard datasets like query log data, and through crowdsourced user studies designed similar to some quantum physics or cognitive science experiments. While the evidence of quantum-like phenomena from standard datasets is not convincing, we find that some of the user studies reveal the quantum-like structure of document judgements. One of the key findings which has implication for IR is the dynamic interactions between the different dimensions of relevance. For example, a user’s judgement of reliability of a document depends significantly on whether they found it understandable or not. Thus, the consideration of one relevance dimension or document feature can provide a context for another dimension, contrary to the current IR models which consider these features to be independent of each other and an objective property of the document. The quantum framework has been especially designed to deal with such scenarios where properties of systems or objects do not exist independent of measurement context. The thesis concludes with suggestions about incorporating quantum mathematical constructs into state-of-art IR algorithms.
September 2006Abstract: Traditional database query languages are based on set theory and crisp first order logic. However, many applications require retrieval-like queries which return result objects associated with a degree value of being relevant to the query. Traditionally, retrieval systems estimate relevance by exploiting hidden object semantics whereas query processing in database systems relies on matching select-conditions with attribute values. Thus, different mechanisms were developed for database and information retrieval systems. In consequence, there is a lack of support for queries involving both retrieval and database search terms. In this work, we develop a unifying framework based on the mathematical formalism of quantum mechanics and quantum logic. Van Rijsbergen already discussed the strong relation between the formalism of quantum mechanics and information retrieval. The goal of this work is to interrelate concepts from database query processing to concepts from quantum mechanics and logic. As result, we obtain a common theory which allows us to incorporate seamlessly retrieval search into traditional database query processing. Exploiting our theoretical results, we introduce the quantum query language QQL. In contrast to competing approaches, our formalism is based on quantum logic.
Recently, Quantum Theory (QT) has been employed to advance the theory of Information Retrieval (IR). Various analogies between QT and IR have been established. Among them, a typical one is applying the idea of photon polarization in IR tasks, e.g., for document ranking and query expansion. In this paper, we aim to further extend this work by constructing a new superposed state of each document in the information need space, based on which we can incorporate the quantum interference idea in query expansion. We then apply the new quantum query expansion model to session search, which is a typical Web search task. Empirical evaluation on the large-scale Clueweb12 dataset has shown that the proposed model is effective in the session search tasks, demonstrating the potential of developing novel and effective IR models based on intuitions and formalisms of QT.