Parsimonious translation models for information retrieval

In the KL divergence framework, the extended language modeling approach has a critical problem of estimating a query model, which is the probabilistic model that encodes the user’s information need. For query expansion in initial retrieval, the translation model had been proposed to involve term co-occurrence statistics. However, the translation model was difficult to apply, because the term co-occurrence statistics must be constructed in the offline time. Especially in a large collection, constructing such a large matrix of term co-occurrences statistics prohibitively increases time and space complexity. In addition, reliable retrieval performance cannot be guaranteed because the translation model may comprise noisy non-topical terms in documents. To resolve these problems, this paper investigates an effective method to construct co-occurrence statistics and eliminate noisy terms by employing a parsimonious translation model. The parsimonious translation model is a compact version of a translation model that can reduce the number of terms containing non-zero probabilities by eliminating non-topical terms in documents. Through experimentation on seven different test collections, we show that the query model estimated from the parsimonious translation model significantly outperforms not only the baseline language modeling, but also the non-parsimonious models.     

An agenda for green information retrieval research

Nowadays we use information retrieval systems and services as part of our many day-to-day activities ranging from a web and database search to searching for various digital libraries, audio and video collections/services, and so on. However, IR systems and services make extensive use of ICT (information and communication technologies) and increasing use of ICT can significantly increase greenhouse gas (GHG, a term used to denote emission of harmful gases in the atmosphere) emissions. Sustainable development, and more importantly environmental sustainability, has become a major area of concern of various national and international bodies and as a result various initiatives and measures are being proposed for reducing the environmental impact of industries, businesses, governments and institutions. Research also shows that appropriate use of ICT can reduce the overall GHG emissions of a business, product or service. Green IT and cloud computing can play a key role in reducing the environmental impact of ICT. This paper proposes the concept of Green IR systems and services that can play a key role in reducing the overall environmental impact of various ICT-based services in education and research, business, government, etc., that are increasingly being reliant on access and use of digital information. However, to date there has not been any systematic research towards building Green IR systems and services. This paper points out the major challenges in building Green IR systems and services, and two different methods are proposed for estimating the energy consumption, and the corresponding GHG emissions, of an IR system or service. This paper also proposes the four key enablers of a Green IR viz. Standardize, Share, Reuse and Green behavior. Further research required to achieve these for building Green IR systems and services are also mentioned.     

An associative search strategy for information retrieval

This paper extends Goffman's Indirect Method of Information Retrieval by suggesting a more flexible search strategy. The suggested strategy in its simplified version is more effective than the Boolean strategy used in existing information retrieval systems and can be implemented at comparable costs. Some of its features might further increase the effectiveness of information retrieval systems.     

The Dilution/Concentration conditions for cross-language information retrieval models

Experimental results of cross-language information retrieval (CLIR) do not indicate why a model fails or how a model could be improved. One basic research question is thus whether it is possible to provide conditions by which one can evaluate any existing or new CLIR strategy analytically and one can improve the design of CLIR models. Inspired by the heuristics in monolingual IR, we introduce in this paper Dilution/Concentration (D/C) conditions to characterize good CLIR models based on direct intuitions under artificial settings. The conditions, derived from first principles in CLIR, generalize the idea of query structuring approach. Empirical results with state-of-the-art CLIR models show that when a condition is not satisfied, it often indicates non-optimality of the method. In general, we find that the empirical performance of a retrieval formula is tightly related to how well it satisfies the conditions. Lastly, we propose, by following the D/C conditions, several novel CLIR models based on the information-based models, which again shows that the D/C conditions are efficient to feature good CLIR models.     

An implicit feedback approach for interactive information retrieval

Searchers can face problems finding the information they seek. One reason for this is that they may have difficulty devising queries to express their information needs. In this article, we describe an approach that uses unobtrusive monitoring of interaction to proactively support searchers. The approach chooses terms to better represent information needs by monitoring searcher interaction with different representations of top-ranked documents. Information needs are dynamic and can change as a searcher views information. The approach we propose gathers evidence on potential changes in these needs and uses this evidence to choose new retrieval strategies. We present an evaluation of how well our technique estimates information needs, how well it estimates changes in these needs and the appropriateness of the interface support it offers. The results are presented and the avenues for future research identified.     

Integration of user profiles: models and experiments in information retrieval

One difficult problem in information retrieval (IR) is the proper interpretation of user queries. It is extremely hard for users to express their information needs in a specific yet exhaustive way. In an effort to alleviate this problem, two theoretical models have been proposed to utilize user characteristics maintained in the form of a user profile. Although the idea of integrating user profiles into an IR system is intuitively appealing, and the models seem viable, no research to date has established a foundation for the roles of user profiles in such a system. Aiming at the investigation of the roles of user profiles, therefore, this study first identifies and extends various query/profile interaction models to provide a ground upon which the investigation can be undertaken. From a continuum of models characterized on the basis of interaction types, metrics, and parameters, nearly 400 models are chosen to investigate the “model space.” New measures are developed based on the notion of user satisfaction/frustration. In addition, three different criteria are used to guide users in making judgments on the quality of retrieved items. Analysis of the data obtained from the experiments shows that, for a wide variety of criteria and metrics, there are always some query/profile interaction models that outperform the query alone model. In addition, preferable characteristics for different criteria are identified in terms of interaction types, parameters, and metrics.     

Contextualization models for XML retrieval

In a hierarchical XML structure, surrounding elements form the context of an XML element. In document-oriented XML, the context is a part of the semantics of the element and augments its textual information. The process of taking the context of the element into account in element scoring is called contextualization. This study extends the concept of contextualization and presents a classification of contextualization models. In an XML collection, elements are of different granularity, i.e. lower level elements are shorter and carry less textual information. Thus, it seems credible that contextualization interacts differently with diverse elements. Even if it is known that contextualization leads to improved effectiveness in element retrieval, the improvement on different granularity levels has not been investigated. This study explores the effect of contextualization on these levels. Further, a parameterized framework for testing contextualization is presented.     

Effective contact recommendation in social networks by adaptation of information retrieval models

We investigate a novel perspective to the development of effective algorithms for contact recommendation in social networks, where the problem consists of automatically predicting people that a given user may wish or benefit from connecting to in the network. Specifically, we explore the connection between contact recommendation and the text information retrieval (IR) task, by investigating the adaptation of IR models (classical and supervised) for recommending people in social networks, using only the structure of these networks.We first explore the use of adapted unsupervised IR models as direct standalone recommender systems. Seeking additional effectiveness enhancements, we further explore the use of IR models as neighbor selection methods, in place of common similarity measures, in user-based and item-based nearest-neighbors (kNN) collaborative filtering approaches. On top of this, we investigate the application of learning to rank approaches borrowed from text IR to achieve additional improvements.We report thorough experiments over data obtained from Twitter and Facebook where we observe that IR models, particularly BM25, are competitive compared to state-of-the art contact recommendation methods. We provide further empirical analysis of the additional effectiveness that can be achieved by the integration of IR models into kNN and learning to rank schemes. Our research shows that the IR models are effective in three roles: as direct contact recommenders, as neighbor selectors in collaborative filtering and as samplers and features in learning to rank.     

An empirical testing of user stereotypes of information retrieval systems

Stereotyping is a technique used in many information systems to represent user groups and/or to generate initial individual user models. However, there has been a lack of evidence on the accuracy of their use in representing users. We propose a formal evaluation method to test the accuracy or homogeneity of the stereotypes that are based on users' explicit characteristics. Using the method, the results of an empirical testing on 11 common user stereotypes of information retrieval (IR) systems are reported. The participants' memberships in the stereotypes were predicted using discriminant analysis, based on their IR knowledge. The actual membership and the predicted membership of each stereotype were compared. The data show that “librarians/IR professionals” is an accurate stereotype in representing its members, while some others, such as “undergraduate students” and “social sciences/humanities” users, are not accurate stereotypes. The data also demonstrate that based on the user's IR knowledge a stereotype can be made more accurate or homogeneous. The results show the promise that our method can help better detect the differences among stereotype members, and help with better stereotype design and user modeling. We assume that accurate stereotypes have better performance in user modeling and thus the system performance.Limitations and future directions of the study are discussed.     

Ontology refinement for improved information retrieval

Ontologies are frequently used in information retrieval being their main applications the expansion of queries, semantic indexing of documents and the organization of search results. Ontologies provide lexical items, allow conceptual normalization and provide different types of relations. However, the optimization of an ontology to perform information retrieval tasks is still unclear. In this paper, we use an ontology query model to analyze the usefulness of ontologies in effectively performing document searches. Moreover, we propose an algorithm to refine ontologies for information retrieval tasks with preliminary positive results.     

Adaptive visualization for exploratory information retrieval

As the volume and breadth of online information is rapidly increasing, ad hoc search systems become less and less efficient to answer information needs of modern users. To support the growing complexity of search tasks, researchers in the field of information developed and explored a range of approaches that extend the traditional ad hoc retrieval paradigm. Among these approaches, personalized search systems and exploratory search systems attracted many followers. Personalized search explored the power of artificial intelligence techniques to provide tailored search results according to different user interests, contexts, and tasks. In contrast, exploratory search capitalized on the power of human intelligence by providing users with more powerful interfaces to support the search process. As these approaches are not contradictory, we believe that they can re-enforce each other. We argue that the effectiveness of personalized search systems may be increased by allowing users to interact with the system and learn/investigate the problem in order to reach the final goal. We also suggest that an interactive visualization approach could offer a good ground to combine the strong sides of personalized and exploratory search approaches. This paper proposes a specific way to integrate interactive visualization and personalized search and introduces an adaptive visualization based search system Adaptive VIBE that implements it. We tested the effectiveness of Adaptive VIBE and investigated its strengths and weaknesses by conducting a full-scale user study. The results show that Adaptive VIBE can improve the precision and the productivity of the personalized search system while helping users to discover more diverse sets of information.     

Query-level loss functions for information retrieval

Many machine learning technologies such as support vector machines, boosting, and neural networks have been applied to the ranking problem in information retrieval. However, since originally the methods were not developed for this task, their loss functions do not directly link to the criteria used in the evaluation of ranking. Specifically, the loss functions are defined on the level of documents or document pairs, in contrast to the fact that the evaluation criteria are defined on the level of queries. Therefore, minimizing the loss functions does not necessarily imply enhancing ranking performances. To solve this problem, we propose using query-level loss functions in learning of ranking functions. We discuss the basic properties that a query-level loss function should have and propose a query-level loss function based on the cosine similarity between a ranking list and the corresponding ground truth. We further design a coordinate descent algorithm, referred to as RankCosine, which utilizes the proposed loss function to create a generalized additive ranking model. We also discuss whether the loss functions of existing ranking algorithms can be extended to query-level. Experimental results on the datasets of TREC web track, OHSUMED, and a commercial web search engine show that with the use of the proposed query-level loss function we can significantly improve ranking accuracies. Furthermore, we found that it is difficult to extend the document-level loss functions to query-level loss functions.     

A logic basis for information retrieval

This paper examines the potential of recent work in artificial intelligence for the development of more effective information retrieval systems. The primary task in this research has been to examine and define the role of an expert system in the domain of bibliographic retrieval. Once such a goal can be described the available knowledge representations and techniques can be evaluated. This paper examines the role of an expert bibliographic retrieval system, examines an artificial intelligence view of information retrieval, and then describes a prototype expert information retrieval system that has been designed and implemented.     

Matching meaning for cross-language information retrieval

This article describes a framework for cross-language information retrieval that efficiently leverages statistical estimation of translation probabilities. The framework provides a unified perspective into which some earlier work on techniques for cross-language information retrieval based on translation probabilities can be cast. Modeling synonymy and filtering translation probabilities using bidirectional evidence are shown to yield a balance between retrieval effectiveness and query-time (or indexing-time) efficiency that seems well suited large-scale applications. Evaluations with six test collections show consistent improvements over strong baselines.     

Dictionary-based techniques for cross-language information retrieval

Cross-language information retrieval (CLIR) systems allow users to find documents written in different languages from that of their query. Simple knowledge structures such as bilingual term lists have proven to be a remarkably useful basis for bridging that language gap. A broad array of dictionary-based techniques have demonstrated utility, but comparison across techniques has been difficult because evaluation results often span only a limited range of conditions. This article identifies the key issues in dictionary-based CLIR, develops unified frameworks for term selection and term translation that help to explain the relationships among existing techniques, and illustrates the effect of those techniques using four contrasting languages for systematic experiments with a uniform query translation architecture. Key results include identification of a previously unseen dependence of pre- and post-translation expansion on orthographic cognates and development of a query-specific measure for translation fanout that helps to explain the utility of structured query methods.     

Conceptual language models for domain-specific retrieval

Over the years, various meta-languages have been used to manually enrich documents with conceptual knowledge of some kind. Examples include keyword assignment to citations or, more recently, tags to websites. In this paper we propose generative concept models as an extension to query modeling within the language modeling framework, which leverages these conceptual annotations to improve retrieval. By means of relevance feedback the original query is translated into a conceptual representation, which is subsequently used to update the query model.     

Novel applications of information retrieval to the storage and management of computer models

Decision-making is often supported by computer-based models. To become a truly valuable corporate resource, such models must be easy to locate, share, and reuse. We describe a technical approach to model management aimed at establishing a database of computer models which can either be reused without modification or modified and composed with other models to assist with novel decisions.We describe a formalism for representing models and discuss various types of queries supported by the formalism. We discuss important research issues that must be addressed for successful application of the formalism to model management. Further, we argue that the field of information retrieval is the natural referent discipline for the study of model storage and retrieval and advance the argument that retrieving computer based models has an important connection to text retrieval based on documents' structure.     

On test collections for adaptive information retrieval

Traditional Cranfield test collections represent an abstraction of a retrieval task that Sparck Jones calls the “core competency” of retrieval: a task that is necessary, but not sufficient, for user retrieval tasks. The abstraction facilitates research by controlling for (some) sources of variability, thus increasing the power of experiments that compare system effectiveness while reducing their cost. However, even within the highly-abstracted case of the Cranfield paradigm, meta-analysis demonstrates that the user/topic effect is greater than the system effect, so experiments must include a relatively large number of topics to distinguish systems’ effectiveness. The evidence further suggests that changing the abstraction slightly to include just a bit more characterization of the user will result in a dramatic loss of power or increase in cost of retrieval experiments. Defining a new, feasible abstraction for supporting adaptive IR research will require winnowing the list of all possible factors that can affect retrieval behavior to a minimum number of essential factors.     

Role-based results redistribution for collaborative information retrieval

We describe a new approach for algorithmic mediation of a collaborative search process. Unlike most approaches to collaborative IR, we are designing systems that mediate explicitly-defined synchronous collaboration among small groups of searchers with a shared information need. Such functionality is provided by first obtaining different rank-lists based on searchers’ queries, fusing these rank-lists, and then splitting the combined list to distribute documents among collaborators according to their roles. For the work reported here, we consider the case of two people collaborating on a search. We assign them roles of Gatherer and Surveyor: the Gatherer is tasked with exploring highly promising information on a given topic, and the Surveyor is tasked with digging further to explore more diverse information. We demonstrate how our technique provides the Gatherer with high-precision results, and the Surveyor with information that is high in entropy.     

Strategies for building distributed information retrieval systems

In this article we discuss the need for distributed information retrieval Systems. A number of possible configurations are presented. A general approach to the design of such systems is discussed. A prototype implementation is described together with the experiences gained from this implementation.