In current systems for browsing image collections, users are presented with sets of thumbnail images arranged in some default order on the screen. We are investigating whether it benefits users to have sets of thumbnails arranged according to their mutual similarity, so images that are alike are placed together. There are, of course, many possible definitions of "similarity": so far we have explored measurements based on low-level visual features, and on the textual captions assigned to the images. Here we describe two experiments, both involving designers as the participants, examining whether similarity-based arrangements of the candidate images are helpful for a picture selection task. Firstly, the two types of similarity-based arrangement were informally compared. Then, an arrangement based on visual similarity was more formally compared with a control of a random arrangement. We believe this work should be of interest to anyone designing a system that involves presenting sets of images to users. Keywords Image retrieval, information visualisation, evaluation.

We consider the problem of organizing and browsing the top ranked portion of the documents returned by an information retrieval system. We study the effectiveness of a document organization in helping a user to locate the relevant material among the retrieved documents as quickly as possible. In this context we examine a set of clustering algorithms and experimentally show that a clustering of the retrieved documents can be significantly more effective than traditional ranked list approach. We also show that the clustering approach can be as effective as the interactive relevance feedback based on query expansion while retaining an important advantage -- it provides the user with a valuable sense of control over the feedback process.

Lighthouse is an on-line interface for a Web-based information retrieval system. It accepts queries from a user, collects the retrieved documents from the search engine, organizes and presents them to the user. The system integrates two known presentations of the retrieved results -- the ranked list and clustering visualization -- in a novel and effective way. It accepts the user's input and adjusts the document visualization accordingly. We give a brief overview of the system. H.3.3 Information Search and Retrieval -- Relevance feedback. H.3.5 Online Information Services -- Web-based services; H.5.2 User Interfaces -- Graphical user interfaces, Screen design; 1. Introduction Locating interesting information on the World Wide Web is the main task of on-line search engines. Such an engine accepts a query from a user and responds with a list of documents or web pages that are considered to be relevant to the query. The pages are ranked by their likelihood of being relevant to the user...

We consider the problem of browsing the top ranked portion of the documents returned by an information retrieval system. We describe an interactive relevance feedback agent that analyzes the inter-document similarities and can help the user to locate the interesting information quickly. We show how such an agent can be designed and improved by using neural networks and reinforcement learning. We demonstrate that its performance significantly exceeds the performance of the traditional relevance feedback approach.

Abstract not found

Retrieval of images from large image archives based solely on their visual similarity to a query image provides an exciting alternative to conventional text-based search. For content-based retrieval images are represented in terms of visual features. The question of how to combine these for similarity computation is typically addressed by eliciting relevance feedback from the user on the retrieved images. We argue in this thesis that the prevailing approach to relevance feedback suffers from three significant shortcomings: firstly, it leaves unsolved the question of how to combine features for the first retrieval; secondly, the advantage of automated content-extraction over manual annotation is greatest for large collections but if the query image is not constrained to come from the indexed collection, content-based retrieval entails imagewise comparisons leading to prohibitive response times; thirdly, users may only have vaguely defined information needs or may change their needs in the course of the interaction. The large majority of relevance feedback techniques are ill-suited for such undirected exploration. We propose a new framework of user interaction that addresses these limitations. It is centred on what we call the NN k idea. The NN k of an image are all those images that are most similar to it under some combination of features. They can be viewed as representatives of the possible

The Boolean semantics of SQL queries cannot adequately capture the “fuzzy ” preferences and “soft ” criteria required in non-traditional data retrieval applications. One way to solve this problem is to add a flavor of “information retrieval ” into database queries by allowing fuzzy query conditions and flexibly supporting grouping and ranking of the query results within the DBMS engine. While ranking is already supported by all major commercial DBMSs natively, support of flexibly grouping is still very limited (i.e., group-by). In this paper, we propose to generalize group-by to enable flexible grouping (clustering specifically) of the query results. Different from clustering in data mining applications, our focus is on supporting efficient clustering of Boolean results generated at query time. Moreover, we propose to integrate ranking and clustering with Boolean conditions, forming a new type of ClusterRank query to allow structured data retrieval. Such an integration is nontrivial in terms of both semantics and query processing. We investigate various semantics of this type of queries. To process such queries, a straightforward approach is to simply glue the techniques developed for ranking-only and clustering-only together. This approach is costly since both ranking and clustering are treated as blocking post-processing tasks upon Boolean query results by existing techniques. We propose a summary-based evaluation method that utilizes bitmap index to seamlessly integrate Boolean conditions, clustering, and ranking. Experimental study shows that our approach significantly outperforms the straightforward one and maintains high clustering quality.

Lighthouse is an on-line interface for a Web-based information retrieval system. It integrates two known presentations of the retrieved results -- the ranked list and clustering visualization -- in a novel and effective way. We describe a working implementation of the system. It accepts queries from a user, collects the retrieved documents from the search engine, organizes and presents them to the user. It is relatively fast and efficient. We also describe some experiments showing that Lighthouse helps the user to locate relevant information much faster than it could be done with the ranked list and can significantly improve the retrieval effectiveness of a search engine. 1 Introduction Locating interesting information on the World Wide Web is the main task of on-line search engines. Such an engine accepts a query from a user and responds with a list of documents or web pages that are considered to be relevant to the query. The pages are ranked by their likelihood of being rele...

News portals gather and organize news articles published daily on the Internet. Typically, news articles are clustered into “events” and each cluster is displayed with a short description of its contents. A particularly interesting choice for describing the contents of a cluster is a machine-generated multi-document summary of the articles in the cluster. Such summaries are informative and help news readers to identify and explore only clusters of interest. Naturally, multi-document clusters and summaries are also valuable to help users navigate the results of keyword-search queries. Unfortunately, current document summarizers are still slow; as a result, search strategies that define document clusters and their multi-document summaries online, in a query-specific manner, are prohibitively expensive. In contrast, search strategies that only return offline, query-independent document clusters are efficient, but might return clusters whose (query-independent) summaries are of little relevance to the queries. In this paper, we present an efficient Hybrid search strategy to address the limitations of fully online and fully offline summarization-aware search approaches. Extensive experiments involving user relevance judgments and real news articles show that the quality of our Hybrid results is high, and that these results are computed in substantially less time than with the fully online strategy. We have implemented our strategy and made it available on the Newsblaster news summarization system, which crawls and summarizes news articles from a variety of web sources on a daily basis.

The visualization of results is a critical component in search engines, and the standard ranked list interface has been a consistently predominant model. The emergence of social media provides a new opportunity to investigate visualization techniques that expose socially derived links between objects to support their exploration. Here we introduce and evaluate network-based visualizations for facilitating the exploration of a Web knowledge space. We developed a force directed network interface to visualize the result sets provided by GiveALink.org, a social bookmarking site. The classifications and tags by users are aggregated to build a social similarity network between bookmarked resources. We administered a user study to evaluate the potential of leveraging such social links in an exploratory search task. During exploration, the similarity links are used to arrange the resources in a semantic layout. Users in our study prefer a hybrid interface combining a conventional ranked list and a two dimensional network map, allowing them to find the same amount of relevant information using fewer queries. This behavior is a direct result of the additional structural information present in the network visualization, which aids them in the exploration of the information space.