Selectively diversifying web search results.



Exploiting query reformulations for web search result diversification.
When a Web user’s underlying information need is not clearly specified from the initial query, an effective approach is to diversify the results retrieved for this query. In this paper, we introduce a novel probabilistic framework for Web search result diversification, which explicitly accounts for the various aspects associated to an underspecified query. In particular, we diversify a document ranking by estimating how well a given document satisfies each uncovered aspect and the extent to which different aspects are satisfied by the ranking as a whole. We thoroughly evaluate our framework in the context of the diversity task of the TREC 2009 Web track. Moreover, we exploit query reformulations provided by three major Web search engines (WSEs) as a means to uncover different query aspects. The results attest the effectiveness of our framework when compared to state-of-the-art diversification approaches in the literature. Additionally, by simulating an upper-bound query reformulation mechanism from official TREC data, we draw useful insights regarding the effectiveness of the query reformulations generated by the different WSEs in promoting diversity.
Finding the right term Retrieving and exploring semantic concepts in astronomical vocabularies.

Blog track research at TREC.
The TREC Blog track aims to explore information seeking behaviour in the blogosphere, by building reusable test collections for blog-related search tasks. Since, its advent in TREC 2006, the Blog track has led to much research in this growing field, and encapsulated cross-pollination from natural language processing research. This paper recaps on the tasks addressed at the TREC Blog track thus far, covering the period 2006- 2009. In particular, we describe the used corpora, the tasks addressed within the track, and the resulting published research. 1
The influence of the document ranking in expert search.
The retrieval effectiveness of the underlying document search component of an expert search engine can have an important impact on the effectiveness of the generated expert search results. In this large-scale study, we perform novel experiments in the context of the document search and expert search tasks of the TREC Enterprise track, to measure the influence that the performance of the document ranking has on the ranking of candidate experts. In particular, we show, using real and simulated document rankings, that while the expert search system performance is related to the relevance of the retrieved documents, surprisingly, it is not always the case that increasing document search effectiveness causes an increase in expert search performance.
A study of selective collection enrichment for enterprise search.

Finding good feedback documents.

Integrating Proximity to Subjective Sentences for Blog Opinion Retrieval.

Selective Application of Query%2DIndependent Features in Web Information Retrieval.

Studying Query Expansion Effectiveness.

Can RDB2RDF Tools Feasibily Expose Large Science Archives for Data Integration%3F.

Browsing Recommendation Based on the Intertemporal Choice Model.

Predicting the Usefulness of Collection Enrichment for Enterprise Search.

Fitting score distribution for blog opinion retrieval.

Is spam an issue for opinionated blog post search%3F

On perfect document rankings for expert search.

On single%2Dpass indexing with MapReduce.

Inverse Document Frequency.
Inverse Document Frequency is a popular measure of a word&#039;s importance. The IDF invariably appears in a host of heuristic measures used in information retrieval. However, so far the IDF has itself been a heuristic. In this paper, we show IDF to be optimal in a principled sense. We show that IDF is the optimal weight of a word with respect to minimization of a Kullback-Leibler distance suitably generalized to nonnegative functions which need not be probability distributions. This optimization problem is closely related to maximum entropy problem. We show that the IDF is the optimal weight associ- ated with a word-feature in an information retrieval setting where we treat each document as the query that retrieves itself. That is, IDF is optimal for doc- ument self-retrieval.
Inverse Document Frequency.
Inverse Document Frequency is a popular measure of a word&#039;s importance. The IDF invariably appears in a host of heuristic measures used in information retrieval. However, so far the IDF has itself been a heuristic. In this paper, we show IDF to be optimal in a principled sense. We show that IDF is the optimal weight of a word with respect to minimization of a Kullback-Leibler distance suitably generalized to nonnegative functions which need not be probability distributions. This optimization problem is closely related to maximum entropy problem. We show that the IDF is the optimal weight associ- ated with a word-feature in an information retrieval setting where we treat each document as the query that retrieves itself. That is, IDF is optimal for doc- ument self-retrieval.
WEB Information Retrieval Models.
Abstract: The research intends to boost the relevance of Web search results by classifying Websnippet into socially constructed hierarchical search concepts, such as the most comprehensive human edited knowledge structure, the Open Directory Project (ODP). The semantic aspects of the search concepts (categories) in the socially constructed hierarchical knowledge repositories are extracted from the associated textual information contributed by societies. The textual information is explored and analyzed to construct a category-document set, which is subsequently employed to represent the semantics of the socially constructed search concepts. Simple API for XML (SAX), a component of JAXP (Java API for XML Processing) is utilized to read in and analyze the two RDF format ODP data files, structure.rdf and content.rdf. kNN, which is trained by the constructed category-document set, is used to categorized the Web search results. The categorized Web search results are then ontologically filtered based on the interactions of Web information seekers. Initial experimental results demonstrate that the proposed approach can improve precision by 23.5%.
Searching for Expertise Experiments with the Voting Model.

Searching for Expertise Experiments with the Voting Model.

Advances in Information Retrieval 
This paper captivates the reader by presenting an overview of the topic detection and tracking (TDT) problem, whose purpose is to organize broadcast news stories by the real-world events that they discuss. The paper describes the research problems considered in all three phases of the TDT and focuses on the solutions developed at CIIR for one of the problems, namely the detection problem--that is, identifying when new topics have appeared in the news stream. The approach discussed is both practical and technically interesting as it presents both new algorithms and modifications of existing IR techniques for the TDT problem
Finding Data Resources in a Virtual Observatory Using SKOS Vocabularies.

Key blog distillation ranking aggregates.

An effective statistical approach to blog post opinion retrieval.

High Quality Expertise Evidence for Expert Search.

Expert Search Evaluation by Supporting Documents.

Analysis of Link Graph Compression Techniques.

On the TREC Blog Track.
The rise of blogging as a new grassroots publishing medium and the many interesting peculiarities that characterise blogs compared to other genres of documents opened up several new interesting research areas in the information retrieval field. The Blog track was introduced in 2006 as part of the renowned Text REtrieval Conference (TREC) evaluation forum, to drive research on the blogosphere and to facilitate experimentation and evaluation of blog search techniques. This paper reports on two years of the Blog track at TREC. We describe the blog search tasks investigated at TREC, and discuss the main lessons we learnt from the track. We conclude the paper with discussions of the broader implications of the Blog track lessons and possible directions for the future, with the aim to uncover and explore the richness of information available in the blogosphere.
Retrieval sensitivity under training using different measures.

Ranking opinionated blog posts using OpinionFinder.

Limits of opinion%2Dfinding baseline systems.

Automatic document prior feature selection for web retrieval.

University of Glasgow at TREC 2008 Experiments in Blog

Voting techniques for expert search.
In TREC 2007, we participate in four tasks of the Blog and Enterprise tracks. We continue experiments using Terrier 1 [14], our modular and scalable Information Retrieval (IR) platform, and the Divergence From Randomness (DFR) framework. In particular, for the Blog track opinion finding task, we propose a statistical term weighting approach to identify opinionated documents. An alternative approach based on an opinion identification tool is also utilised. Overall, a 15 % improvement over a non-opinionated baseline is observed in applying the statistical term weighting approach. In the Expert Search task of the Enterprise track, we investigate the use of proximity between query terms and candidate name occurrences in documents. 1.
A syntactically%2Dbased query reformulation technique for information retrieval.

Voting techniques for expert search.
In TREC 2007, we participate in four tasks of the Blog and Enterprise tracks. We continue experiments using Terrier 1 [14], our modular and scalable Information Retrieval (IR) platform, and the Divergence From Randomness (DFR) framework. In particular, for the Blog track opinion finding task, we propose a statistical term weighting approach to identify opinionated documents. An alternative approach based on an opinion identification tool is also utilised. Overall, a 15 % improvement over a non-opinionated baseline is observed in applying the statistical term weighting approach. In the Expert Search task of the Enterprise track, we investigate the use of proximity between query terms and candidate name occurrences in documents. 1.
30th European Conference in Information Retrieval %28ECIR 2008%29.

Parameter sensitivity in the probabilistic model for ad%2Dhoc retrieval.

Expertise drift and query expansion in expert search.

Performance Comparison of Clustered and Replicated Information Retrieval Systems.

Multinomial Randomness Models for Retrieval with Document Fields.
Abstract. Document fields, such as the title or the headings of a document, offer a way to consider the structure of documents for retrieval. Most of the proposed approaches in the literature employ either a linear combination of scores assigned to different fields, or a linear combination of frequencies in the term frequency normalisation component. In the context of the Divergence From Randomness framework, we have a sound opportunity to integrate document fields in the probabilistic randomness model. This paper introduces novel probabilistic models for incorporating fields in the retrieval process using a multinomial randomness model and its information theoretic approximation. The evaluation results from experiments conducted with a standard TREC Web test collection show that the proposed models perform as well as a state-of-the-art field-based weighting model, while at the same time, they are theoretically founded and more extensible than current field-based models. 1
Using Relevance Feedback in Expert Search.
Abstract. In Enterprise settings, expert search is considered an important task. In this search task, the user has a need for expertise- for instance, they require assistance from someone about a topic of interest. An expert search system assists users with their “expertise need ” by suggesting people with relevant expertise to the topic of interest. In this work, we apply an expert search approach that does not explicitly rank candidates in response to a query, but instead implicitly ranks candidates by taking into account a ranking of document with respect to the query topic. Pseudo-relevance feedback, aka query expansion, has been shown to improve retrieval performance in adhoc search tasks. In this work, we investigate to which extent query expansion can be applied in an expert search task to improve the accuracy of the generated ranking of candidates. We define two approaches for query expansion, one based on the initial of ranking of documents for the query topic. The second approach is based on the final ranking of candidates. The aims of this paper are two-fold. Firstly, to determine if query expansion can be successfully applied in the expert search task, and secondly, to ascertain if either of the two forms of query expansion can provide robust, improved retrieval performance. We perform a thorough evaluation contrasting the two query expansion approaches in the context of the TREC 2005 and 2006 Enterprise tracks. 1
Setting Per%2Dfield Normalisation Hyper%2Dparameters for the Named%2DPage Finding Search Task.

Combination of Document Priors in Web Information Retrieval.
Query-independent features (also called document priors), such as the number of incoming links to a document, its Page-Rank, or the type of its associated URL, have been successfully integrated into Web Information Retrieval systems in order to enhance the retrieval effectiveness. The combination of several document priors could further enhance the retrieval performance. However, most current combination of priors approaches are based on heuristics, and often ignore the possible dependence between the document priors. In this paper, we present a novel and robust method for conditionally combining document priors in a principled way. The approach adjusts the distribution of document priors for one source of evidence according to the distribution of document priors for other sources of evidence. We investigate the retrieval performance attainable by our combination of priors method, in comparison to the use of single priors and to a heuristic combination of document priors method, which assumes that document priors are independent. Furthermore, we investigate how sensitive the proposed method is to the training data. Using two standard Web test collections, including the large-scale.GOV2 test collection, we find that some of the document priors used in our experiments, have a considerably high correlation, suggesting that the dependency between documents priors should indeed be taken into account. Through extensive experiments on these two large-scale collections, we observe that our proposed conditional combination method is overall effective and robust. 1
Light Syntactically%2DBased Index Pruning for Information Retrieval.
Abstract. Most index pruning techniques eliminate terms from an index on the basis of the contribution of those terms to the content of the documents. We present a novel syntactically-based index pruning technique, which uses exclusively shallow syntactic evidence to decide upon which terms to prune. This type of evidence is document-independent, and is based on the assumption that, in a general collection of documents, there exists an approximately proportional relation between the frequency and content of ‘blocks of parts of speech ’ (POS blocks) [5]. POS blocks are fixed-length sequences of nouns, verbs, and other parts of speech, extracted from a corpus. We remove from the index, terms that correspond to low-frequency POS blocks, using two different strategies: (i) considering that low-frequency POS blocks correspond to sequences of content-poor words, and (ii) considering that low-frequency POS blocks, which also contain ‘non content-bearing parts of speech’, such as prepositions for example, correspond to sequences of contentpoor words. We experiment with two TREC test collections and two statistically different weighting models. Using full indices as our baseline, we show that syntactically-based index pruning overall enhances retrieval performance, in terms of both average and early precision, for light pruning levels, while also reducing the size of the index. Our novel low-cost technique performs at least similarly to other related work, even though it does not consider document-specific information, and as such it is more general. 1
Combination of Document Priors in Web Information Retrieval.
Query-independent features (also called document priors), such as the number of incoming links to a document, its Page-Rank, or the type of its associated URL, have been successfully integrated into Web Information Retrieval systems in order to enhance the retrieval effectiveness. The combination of several document priors could further enhance the retrieval performance. However, most current combination of priors approaches are based on heuristics, and often ignore the possible dependence between the document priors. In this paper, we present a novel and robust method for conditionally combining document priors in a principled way. The approach adjusts the distribution of document priors for one source of evidence according to the distribution of document priors for other sources of evidence. We investigate the retrieval performance attainable by our combination of priors method, in comparison to the use of single priors and to a heuristic combination of document priors method, which assumes that document priors are independent. Furthermore, we investigate how sensitive the proposed method is to the training data. Using two standard Web test collections, including the large-scale.GOV2 test collection, we find that some of the document priors used in our experiments, have a considerably high correlation, suggesting that the dependency between documents priors should indeed be taken into account. Through extensive experiments on these two large-scale collections, we observe that our proposed conditional combination method is overall effective and robust. 1
Incorporating term dependency in the dfr framework.
Term dependency, or co-occurrence, has been studied in language modelling, for instance by Metzler &amp; Croft [3] who showed that retrieval performance could be significantly enhanced using term dependency information. In this work, we show how term dependency can be modelled within the Divergence From Randomness (DFR) framework. We evaluate our term dependency model on the two adhoc retrieval tasks using the TREC.GOV2 Terabyte collection. Furthermore, we examine the effect of varying the term dependency window size on the retrieval performance of the proposed model. Our experiments show that term dependency can indeed be successfully incorporated within the DFR framework.
Extending Weighting Models with a Term Quality Measure.
Abstract. Weighting models use lexical statistics, such as term frequencies, to derive term weights, which are used to estimate the relevance of a document to a query. Apart from the removal of stopwords, there is no other consideration of the quality of words that are being ‘weighted’. It is often assumed that term frequency is a good indicator for a decision to be made as to how relevant a document is to a query. Our intuition is that raw term frequency could be enhanced to better discriminate between terms. To do so, we propose using non-lexical features to predict the ‘quality ’ of words, before they are weighted for retrieval. Specifically, we show how parts of speech (e.g. nouns, verbs) can help estimate how informative a word generally is, regardless of its relevance to a query/document. Experimental results with two standard TREC 1 collections show that integrating the proposed term quality to two established weighting models enhances retrieval performance, over a baseline that uses the original weighting models, at all times. 1
Overview of the TREC 2007 Blog Track.


University of Glasgow at TREC 2007 Experiments in Blog and Enterprise Tracks with Terrier.
In TREC 2007, we participate in four tasks of the Blog and Enterprise tracks. We continue experiments using Terrier 1 [14], our modular and scalable Information Retrieval (IR) platform, and the Divergence From Randomness (DFR) framework. In particular, for the Blog track opinion finding task, we propose a statistical term weighting approach to identify opinionated documents. An alternative approach based on an opinion identification tool is also utilised. Overall, a 15 % improvement over a non-opinionated baseline is observed in applying the statistical term weighting approach. In the Expert Search task of the Enterprise track, we investigate the use of proximity between query terms and candidate name occurrences in documents. 1.

Performance analysis of distributed information retrieval architectures using an improved network simulation model.
Abstract. In this study, we present the analysis of the interconnection network of a distributed Information Retrieval (IR) system, by simulating a switched network versus a shared access network. The results show that the use of a switched network improves the performance, especially in a replicated system because the switched network prevents the saturation of the network, particularly when using a large number of query servers. 1
Combining fields for query expansion and adaptive query expansion.

Examining the Content Load of Part of Speech Blocks for Information Retrieval.

Voting for candidates adapting data fusion techniques for an expert search task.

Combining fields in known%2Ditem email search.
Emails are examples of structured documents with various fields. These fields can be exploited to enhance the retrieval effectiveness of an Information Retrieval (IR) system that searches mailing list archives. In recent experiments of the TREC2005 Enterprise track, various fields were applied to varying degrees of success by the participants. In this work, using a field-based weighting model, we investigate the retrieval performance attainable by each field, and examine when fields evidence should be combined or not.
Searching for expertise using the terrier platform.

Overview of the TREC 2006 Blog Track.
The rise on the Internet of blogging, the creation of journal-like web page logs, has created a highly dynamic subset of the World Wide Web that evolves and responds to real-world events. Indeed, blogs (or weblogs) have recently emerged as a new grassroots publishing medium. The so-called blogosphere (the collection of blogs on
University of Glasgow at TREC 2006 Experiments in Terabyte and Enterprise Tracks with Terrier.
In TREC 2006, we participate in three tasks of the Terabyte and Enterprise tracks. We continue experiments using Terrier 1, our modular and scalable Information Retrieval (IR) platform. Furthering our research into the Divergence From Randomness (DFR) framework of weighting models, we introduce two new effective and low-cost models, which combine evidence from document structure and capture term dependence and proximity, respectively. Additionally, in the Terabyte track, we improve on our query expansion mechanism on fields, presented in TREC 2005, with a new and more refined technique, which combines evidence in a linear, rather than uniform, way. We also introduce a novel, low-cost syntacticallybased noise reduction technique, which we flexibly apply to both the queries and the index. Furthermore, in the Named Page Finding task, we present a new technique for combining query-independent evidence, in the form of prior probabilities. In the Enterprise track, we test our new voting model for expert search. Our experiments focus on the need for candidate length normalisation, and on how retrieval performance can be enhanced by applying retrieval techniques to the underlying ranking of documents. 1.
Query reformulation using automatically generated query concepts from a document space.

A decision mechanism for the selective combination of evidence in topic distillation.
The combination of evidence can increase retrieval effectiveness. In this paper, we investigate the effectiveness of a decision mechanism for the selective combination of evidence for Web Information Retrieval and particularly for topic distillation. We introduce two measures of a query’s broadness and use them to select an appropriate combination of evidence for each query. The results from our experiments show that there is a statistically significant association between the output of the decision mechanism and the relative effectiveness of the different combinations of evidence. Moreover, we show that the proposed methodology can be applied in an operational setting, where relevance information is not available, by setting the decision mechanism’s thresholds automatically.
Query performance prediction.
We present a novel framework for the query-performance prediction task. That is, estimating the effectiveness of a search performed in response to a query in lack of relevance judgments. Our approach is based on using statistical decision theory for estimating the utility that a document ranking provides with respect to an information need expressed by the query. To address the uncertainty in inferring the information need, we estimate utility by the expected similarity between the given ranking and those induced by relevance models; the impact of a relevance model is based on its presumed representativeness of the information need. Specific query-performance predictors instantiated from the framework substantially outperform state-of-the-art predictors over five TREC corpora.
Applying Light Natural Language Processing to Ad%2DHoc Cross Language Information Retrieval.

University of Glasgow at WebCLEF 2005 Experiments in Per%2DField Normalisation and Language Specific Stemming.

Term Frequency Normalisation Tuning for BM25 and DFR Models.
Abstract. The term frequency normalisation parameter tuning is a crucial issue in information retrieval (IR), which has an important impact on the retrieval performance. The classical pivoted normalisation approach suffers from the collection-dependence problem. As a consequence, it requires relevance assessment for each given collection to obtain the optimal parameter setting. In this paper, we tackle the collection-dependence problem by proposing a new tuning method by measuring the normalisation effect. The proposed method refines and extends our methodology described in [7]. In our experiments, we evaluate our proposed tuning method on various TREC collections, for both the normalisation 2 of the Divergence From Randomness (DFR) models and the BM25’s normalisation method. Results show that for both normalisation methods, our tuning method significantly outperforms the robust empirically-obtained baselines over diverse TREC collections, while having a marginal computational cost. 1
Terrier Information Retrieval Platform.
There is increasing interest in improving the robustness of IR systems,  i.e. their effectiveness on difficult queries. A system is robust when it achieves  both a high Mean Average Precision (MAP) value for the entire set of topics and  a significant MAP value over its worst X topics (MAP(X)). It is a well known  fact that Query Expansion (QE) increases global MAP but hurts the performance  on the worst topics. A selective application of QE would thus be a natural answer  to obtain a more robust retrieval system.
Network Analysis for Distributed Information Retrieval Architectures.
Abstract. In this study, we present the analysis of the interconnection network of a distributed Information Retrieval (IR) system, by simulating a switched network versus a shared access network. The results show that the use of a switched network improves the performance, especially in a replicated system because the switched network prevents the saturation of the network, particularly when using a large number of query servers. 1
A study of the dirichlet priors for term frequency normalisation.
applied to the smoothing technique of the language modeling approach. In this paper, we apply the Dirichlet Priors to the term frequency normalisation of the classical BM25 probabilistic model and the Divergence from Randomness PL2 model. The contributions of this paper are twofold. First, through extensive experiments on four TREC collections, we show that the newly generated models, to which the Dirichlet Priors normalisation is applied, provide robust and effective performance. Second, we propose a novel theoreticallydriven approach to the automatic parameter tuning of the Dirichlet Priors normalisation. Experiments show that this tuning approach optimises the retrieval performance of the newly generated Dirichlet Priors-based weighting models.
University of Glasgow at TREC 2005 Experiments in Terabyte and Enterprise Tracks with Terrier.
With our participation in TREC 2005, we continue experiments using Terrier, a modular and scalable Information  Retrieval (IR) framework, in 4 tasks from the Terabyte and Enterprise tracks. In the Terabyte track, we investigate  new Divergence From Randomness weighting models, and a novel query expansion approach that can take into  account various document fields, namely content, title and anchor text. In addition, we test a new selective query  expansion mechanism which determines the appropriateness of using query expansion on a per-query basis,  using statistical information from a low-cost query performance predictor. In the Enterprise track, we investigate  combining document fields evidence with other information occurring in an Enterprise setting. In the email  known item task, we also investigate temporal and thread priors suitable for email search. In the expert search  task, for each candidate, we generate profiles of expertise evidence from the W3C collection. Moreover, we  propose a model for ranking these candidate profiles in response to a query.
A case study of distributed information retrieval architectures to index one terabyte of text.

Dempster%2DShafer Theory for a Query%2DBiased Combination of Evidence on the Web.
Abstract. This paper reports on a large-scale experiment for the evaluation of a formal query-biased combination of evidence mechanism. We use the Dempster-Shafer theory of evidence to combine optimally results obtained by content and link analyses on the Web. The query-biased mechanism is based on the query scope, a measure of the query specificity. The query scope is defined using a probabilistic propagation mechanism on top of the hierarchical structure of concepts provided by WordNet. We use two standard Web test collections and two different link analysis approaches. The results show that the proposed approach could improve the retrieval effectiveness. Keywords: Combination of content and link analysis, Query-Biased combination of evidence, Dempster-Shafer theory of evidence, Web Information Retrieval, Query scope
The Static Absorbing Model for the Web.
This working paper reviews some di#erent ideas in link-based analysis for search. First, results about static ranking of web pages based on the so called randomsurfer model are reviewed and presented in a unified framework. Second, a topic-based hubs and authorities model using a discrete component method (a variant of ICA and PCA) is developed, and illustrated on the 500,000 page English language Wikipedia collection. Third, a proposal is presented to the community for a Links/Ranking consortium extracted from the Web Intelligence paper Opportunities from Open Source Search.
ACM SIGIR workshop on mathematical%2Fformal methods in information retrieval MF%2FIR 2005.

Gauging Students%27 Understanding Through Interactive Lectures.

The University of Glasgow at CLEF 2004 French Monolingual Information Retrieval with Terrier.

Performance Analysis of Distributed Architectures to Index One Terabyte of Text.
Abstract. We simulate different architectures of a distributed Information Retrieval system on a very large Web collection, in order to work out the optimal setting for a particular set of resources. We analyse the effectiveness of a distributed, replicated and clustered architecture using a variable number of workstations. A collection of approximately 94 million documents and 1 terabyte of text is used to test the performance of the different architectures. We show that in a purely distributed architecture, the brokers become the bottleneck due to the high number of local answer sets to be sorted. In a replicated system, the network is the bottleneck due to the high number of query servers and the continuous data interchange with the brokers. Finally, we demonstrate that a clustered system will outperform a replicated system if a large number of query servers is used, mainly due to the reduction of the network load. 1
Construction of Query Concepts in a Document Space Based on Data Mining Techniques.

Selective Combination of Evidence for Topic Distillation using Document and Aggregate%2Dlevel Information.
The combination of evidence for Information Retrieval has been studied extensively in order to increase effectiveness. In this paper, we study the selective application of different retrieval approaches on a per-query basis for Web Information Retrieval. Our methodology is based on the assumption that not all queries benefit from a uniform retrieval approach. In order to select the most appropriate retrieval approaches for each query, we use evidence from the hyperlink structure of the retrieved documents and also from the distribution of aggregates, that is the groups of documents from the same domain. Our experimental results show that it is possible to obtain important improvements in retrieval effectiveness by using simple statistical decision mechanisms on a per-query basis.
A Query%2Dbased Pre%2Dretrieval Model Selection Approach to Information Retrieval.
In this paper, we propose a query-based pre-retrieval approach to the model selection problem,  which automatically selects the best-performing retrieval model before the retrieval process  takes place. In this approach, the queries are clustered according to their statistics and the bestperforming  retrieval model is associated to each cluster. For a given new query, we assign the  closest cluster to the query, and then we apply the model associated to the cluster. We evaluate  the model selection approach on the disk1&amp;2 of the TREC collections. The results show that  our model selection approach achieves stable performance, which could outperform the use of  the optimal retrieval model indi#erently for each query. The results also show that, interestingly,  a retrieval model provides consistent performance for queries belonging to the same cluster.
Usefulness of hyperlink structure for query%2Dbiased topic distillation.
In this paper, we introduce an information theoretic method for estimating the usefulness of the hyperlink structure induced from the set of retrieved documents. We evaluate the effectiveness of this method in the context of an optimal Bayesian decision mechanism, which selects the most appropriate retrieval approaches on a perquery basis for two TREC tasks. The estimation of the hyperlink structure’s usefulness is stable when we use different weighting schemes, or when we employ sampling of documents to reduce the computational overhead. Next, we evaluate the effectiveness of the hyperlink structure’s usefulness in a realistic setting, by setting the thresholds of a decision mechanism automatically. Our results show that improvements over the baselines are obtained.
Discovery of aggregate usage profiles based on clustering information needs.

University of Glasgow at TREC 2004 Experiments in Web
With our participation in TREC2004, we test Terrier, a modular and scalable Information Retrieval framework, in three tracks. For the mixed query task of the Web track, we employ a decision mechanism for selecting appropriate retrieval approaches on a per-query basis. For the robust track, in order to cope with the poorlyperforming queries, we use two pre-retrieval performance predictors and a weighting function recommender mechanism. We also test a new training approach for the automatic tuning of the term frequency normalisation parameters. In the Terabyte track, we employ a distributed version of Terrier and test the effectiveness of techniques, such as using the anchor text, pseudo query expansion and selecting different weighting models for each query. 1
Inferring Query Performance Using Pre%2Dretrieval Predictors..
Abstract. The prediction of query performance is an interesting and important issue in Information Retrieval (IR). Current predictors involve the use of relevance scores, which are time-consuming to compute. Therefore, current predictors are not very suitable for practical applications. In this paper, we study a set of predictors of query performance, which can be generated prior to the retrieval process. The linear and non-parametric correlations of the predictors with query performance are thoroughly assessed on the TREC disk4 and disk5 (minus CR) collections. According to the results, some of the proposed predictors have significant correlation with query performance, showing that these predictors can be useful to infer query performance in practical applications. 1
University of Glasgow at TREC 2004 Experiments in Web
With our participation in TREC2004, we test Terrier, a modular and scalable Information Retrieval framework, in three tracks. For the mixed query task of the Web track, we employ a decision mechanism for selecting appropriate retrieval approaches on a per-query basis. For the robust track, in order to cope with the poorlyperforming queries, we use two pre-retrieval performance predictors and a weighting function recommender mechanism. We also test a new training approach for the automatic tuning of the term frequency normalisation parameters. In the Terabyte track, we employ a distributed version of Terrier and test the effectiveness of techniques, such as using the anchor text, pseudo query expansion and selecting different weighting models for each query. 1
Distribution of relevant documents in domain%2Dlevel aggregates for topic distillation.
In this paper, we study the distribution of relevant documents in aggregates, formed by grouping the retrieved documents according to their domain. For each aggregate, we take into account its size, and a measure of the correlation between its incoming and outgoing hyperlinks. We report on a preliminary experiment with two TREC topic distillation tasks, where we find that larger aggregates, or those aggregates with correlated hyperlinks, are more likely to contain relevant documents. This result shows that the distribution of domain-level aggregates is potentially useful for finding relevant documents.
A study of parameter tuning for term frequency normalization.

A Weighting Scheme for Star%2DGraphs.

Information Theoretic Query Context Detection for Web Information Retrieval.

A Utility%2DOriented Hyperlink Analysis Model for the Web.

University of Glasgow at the Robust Track%2D A Query%2Dbased Model Selection Approach for the Poorly%2DPerforming Queries.

University of Glasgow at the Web Track Dynamic Application of Hyperlink Analysis using the Query Scope.
This year, our participation to the Web track aims at combining dynamically evidence from both content and hyperlink analysis. To this end, we introduce a decision mechanism based on the so-called query scope concept. For the topic distillation task, we find that the use of anchor text increases precision significantly over contentonly retrieval, a result that contrasts with our TREC11 findings. Using the query scope, we show that a selective application of hyperlink analysis, or URL-based scores, is effective for the more generic queries, improving the overall precision. In fact, our most effective runs use the decision mechanism and outperform significantly the content and anchor text retrieval. For the known item task, we employ the query scope in order to distinguish the named page queries from the home page queries, obtaining results close to the content and anchor text baseline. 1
Report on ACM SIGIR workshop on %22semantic web%22 SWIR 2003.

Building Bilingual Dictionaries from Parallel Web Documents.
In this paper we describe a system for automatically constructing a  bilingual dictionary for cross-language information retrieval applications. We  describe how we automatically target candidate parallel documents, filter the  candidate documents and process them to create parallel sentences. The parallel  sentences are then automatically translated using an adaptation of the EMIM  technique and a dictionary of translation terms is created. We evaluate our  dictionary using human experts. The evaluation showed that the system  performs well. In addition the results obtained from automatically-created  corpora are comparable to those obtained from manually created corpora of  parallel documents. Compared to other available techniques, our approach has  the advantage of being simple, uniform, and easy-to-implement while providing  encouraging results.
University of Glasgow at the Web Track of TREC 2002.
The aim of our participation in the topic distillation and the named page finding tasks of the Web track is the evaluation of a well-founded modular probabilistic framework for Web Information Retrieval, which integrates content and link analyses. The link analysis component of the framework employs a new probabilistic approach, called the Absorbing Model, for calculating a measure of popularity for documents induced from the Web graph.
Using Hierarchical Clustering and Summarisation Approaches for Web Retrieval Glasgow at the TREC 2002 Interactive Track.
Introduction and Motivation  Current search engines axe typified as having a lack of precision, coupled with an elongated ranked list style of result presentation. When combined, these factors make relevant data extraction increasingly complex. The main investigation of our participation in the Interactive Track of TREC 2002 is to assess the effectiveness of new visualisation techniques for displaying the results of search engines.  Our current system, provisionally named HuddleSearch, uses a newly developed clustering algorithm, which dynamically organises the relevant documents into a traversable hierarchy of general to more-specific cluster categories. We have extended our TREC-10 summarisation tool to also allow the summarisation of multiple documents; whereby a summary paints a caricature of the contents of a cluster, rather than an individual document, thus allowing the user to provisionally judge a cluster&#039;s relevance prior to viewing its contents. The interaction between the us
Evaluating Summarisation Technologies A Task Oriented Approach.
. This paper presents a novel task-oriented approach for the evaluation  of automatic text summarisation systems. Evaluation of systems has traditionally  been a troublesome area in summarisation research. We propose a scheme  that evaluates three existing systems by determining their relative effectiveness  in an interactive search task, under conditions that approximate the intended  use of the systems.  1. 
Toward a User%2DCentered Image Retrieval System.
Image ranking is a task that involves different parameters. They depend on the intrinsic characteristics of an image, but also on the indexing language used for representing its semantic content. We developed a weighting model that combines these parameters in a general scheme. Finding the best balance between the parameters is not straightforward. Different parameter combinations leads to different rankings, which may be more or less accepted by the users. In this paper, we choose a set of test queries and present the impact of the parameters on the rank of each image. Different combinations are discussed, and the best combination is specified. For the evaluation, we follow a user-oriented approach, and compare the ranking provided by each parameter combination to the ranking given by human judgment. This is a step toward a user-centered image retrieval system, which will dynamically adapt to the user&#039;s profile and preferences.  1 Introduction  Images constitute a complex type of medi...
Ranking digital images using combination if evidences.

Conceptual Graphs and First Order Logic.
this paper, conceptual graphs consist in a convivial graphical representation of logic
A Flexible Weighting Scheme for Multimedia Documents.

Effective and Efficient Relational Query Processing Using Conceptual Graphs.
In a recent paper (see Ounis &amp; Huibers&#039;s work), a logical relational framework for information retrieval is presented,  which emphasizes the importance of relations for accurate indexing, as well as the need to provide relation  treatments for effective retrieval. Most knowledge representation formalisms support relational indexing. However,  the majority of them do not fully allow relation treatment. Conceptual graphs, though offering the richness needed  to use relations in the indexing of complex and highly structured documents, do not provide further relation-based  processing. In this paper, we follow the above relational framework in the case of conceptual graphs, such that we  can take into account relation properties in the retrieval process. The approach leads to a sound extension which  preserves the semantics of this formalism. In its implementation, an important goal was to obtain a workable system.
Modeling%2C Indexing and Retrieving Images Using Conceptual Graphs.
. When dealing with the complexity of an image as part of the indexing  process, keywords are not sufficient to obtain an index that is a faithful representation  of the image content. We propose to use the conceptual graphs formalism as  the indexing language, which allows to use not only keywords, but also relations  between them. The obtained indexes are more precise, and retrieval effectiveness  is thus improved. Our paper presents a system that provides a computer-assisted  image indexing process, which is performed according to a formal image model.  The result of the indexing process, which is a set of conceptual graphs, is then organized  so that to improve retrieval execution times. Our image retrieval system,  called RELIEF, is implemented on an object-oriented DBMS and is available on  the Web. It ensures the management of an image test collection and gives good  results, with respect to both execution time and quality of answers.  1 Introduction: Towards Precision-Oriented ...
A Promising Retrieval Algorithm for Systems Based on the Conceptual Graphs Formalism.

RELIEF Combining Expressiveness and Rapidity into a Single System.
This paper constitutes a proposal for an efficient and effective logical information retrieval system. Following a relational indexing approach, which is in our opinion a necessity to cope with the emerging applications such as those based on multimedia, we use the conceptual graphs formalism as our indexing language. This choice allows for relational indexing support and captures all the useful properties of the logical information retrieval model, in a workable system. First order logic and standard information retrieval techniques are combined together, to the same effect: obtaining an expressive system, able to accurately handle complex documents, improve retrieval effectiveness, and achieve good time performance. Experimentations on an image test collection, within a system available on the Web, provide an illustration of the role that logic may have in the future development of information retrieval systems.  1 Introduction  The emergence of new applications, such as those based ...
Towards a Fast Precision%2DOriented Image Retrieval System.

A Logical Relational Approach for Information Retrieval Indexing.
In a relational indexing approach (see e.g. Farradane&#039;s work), information is carried by a fixed set of relationship types over an underlying set of terms. The idea is that the essence of the meaning of information is encapsulated in the relationships between terms. The importance of relationships is now widely recognized within many fields such as relational databases and knowledge representation formalisms. These fields have substantially improved our understanding of relationships and the problems involved in trying to formalize them. However, although those relationships can be correctly represented by almost all the well-known formalisms in such fields, they are not exploited as much as the objects by concrete operations. In information retrieval, previous attempts at managing relationships have mainly addressed structural aspects, and exclude the manipulation of index expressions by relational operations. This paper suggests a prime use of the relation properties through ...
An Extended Inverted File Approach for Information Retrieval.

Using Conceptual Graphs in a Multifaceted Logical Model for Information Retrieval.
. In 1986, van Rijsbergen promotes a principle in which the matching  between a query and a document is founded on a form of uncertain logical inference.  His formulation was deliberately abstract, no attempt was made to specify  how such a principle might be implemented through an effective logic and under  a particular semantics. In classical logics inference which is often associated with  logical implication is too strict to be useful in the matching process. To overcome  this problem, the possible worlds semantics seems to be a good way to formalize  the uncertain inferences needed in an information retrieval system. Thus, in 1988  Nie initiated a work which has as its basic assumption the consideration that a  document can be associated to a possible world. We think however that a more  refined approach must be adopted especially when complex information are considered.  It is based on the notion of facet, which allows to consider a document  under several points of view, every o...
Conceptual Graph Aboutness.
. In this paper we present an information retrieval model based on conceptual  graphs named ELEN [1]. In ELEN a conceptual graph is a representative of  the information inherent in the document and query. Graph operators can be used  to determine whether a graph is a generalisation of another graph, in which case  the information carried by the graph is said to be about the information carried  by the generalisation of the graph. In this article we want to compare this conceptual  graph aboutness with aboutness derivations of other information retrieval  models. The aboutness derivations are first mapped to a situated logic framework.  Within this framework, the axioms that drive the retrieval process can be filtered  out. The retrieval mechanism can then be compared with other approaches according  to which axioms they are governed by. Using this method we can show that  ELEN is indeed closely related to other classical aboutness derivations, and that in  some aspects it has even more...
