Robert Villa

The effectiveness of query-specific hierarchic clustering in information retrieval

Hierarchic document clustering has been widely applied to information retrieval (IR) on the grounds of its potential improved effectiveness over inverted file search (IFS). However, previous research has been inconclusive as to whether clustering does bring improvements. In this paper we take the view that if hierarchic clustering is applied to search results (query-specific clustering), then it has the potential to increase the retrieval effectiveness compared both to that of static clustering and of conventional IFS. We conducted a number of experiments using five document collections and four hierarchic clustering methods. Our results show that the effectiveness of query-specific clustering is indeed higher, and suggest that there is scope for its application to IR.

A multi-criteria content-based filtering system

In this paper we present a novel filtering system, based on a new model which reshapes the aims of content-based filtering. The filtering system has been developed within the EC project PENG, aimed at providing news professionals, such as journalists, with a system supporting both filtering and retrieval capabilities. In particular, we suggest that in tackling the problem of information overload, it is necessary for filtering systems to take into account multiple aspects of incoming documents in order to estimate their relevance to a users profile, and in order to help users better understand documents, as distinct from solely attempting to either select relevant material from a stream, or block inappropriate material. Aiming to so this, a filtering model based on multiple criteria has been defined, based on the ideas gleamed in the project requirements stage. The filtering model is briefly described in this paper.

Simulated Testing of an Adaptive Multimedia Information Retrieval System

The Semantic Gap is considered to be a bottleneck in image and video retrieval. One way to increase the communication between user and system is to take advantage of the users action with a system, e.g. to infer the relevance or otherwise of a video shot viewed by the user. In this paper we introduce a novel video retrieval system and propose a model of implicit information for interpreting the users actions with the interface. The assumptions on which this model was created are then analysed in an experiment using simulated users based on relevance judgements to compare results of explicit and implicit retrieval cycles. Our model seems to enhance retrieval results. Results are presented and discussed in the final section.

Diversity, Assortment, Dissimilarity, Variety: A Study of Diversity Measures Using Low Level Features for Video Retrieval

In this paper we present a number of methods for re-ranking video search results in order to introduce diversity into the set of search results. The usefulness of these approaches is evaluated in comparison with similarity based measures, for the TRECVID 2007 collection and tasks [11]. For the MAP of the search results we find that some of our approaches perform as well as similarity based methods. We also find that some of these results can improve the P@N values for some of the lower N values. The most successful of these approaches was then implemented in an interactive search system for the TRECVID 2008 interactive search tasks. The responses from the users indicate that they find the more diverse search results extremely useful.

How is a search system used in work task completion

Typically studies of information retrieval and interactive information retrieval concentrate on the identification of relevant items. In this study, rather than stop at finding relevant items, we considered how people use a search system in the completion of a broader work task. To conduct the study, we created 12 tasks that required multiple queries and document views in order to find enough information to complete the task. A total of 381 people completed three tasks each in a laboratory setting using the wikiSearch system that was embedded into WiIRE. Results found that two-thirds of time spent on the task was spent after finding a relevant set of documents sufficient for task completion, and that time was mainly spent reviewing documents that had already been retrieved. Findings suggest that an open-source information retrieval system, such as Lucene, was adequate for this task. However, the ultimate challenge will be in building useful systems that aid the user in extracting, interpreting and analysing information to achieve work task completion.

Simulated evaluation of faceted browsing based on feature selection

In this paper we explore the limitations of facet based browsing which uses sub-needs of an information need for querying and organising the search process in video retrieval. The underlying assumption of this approach is that the search effectiveness will be enhanced if such an approach is employed for interactive video retrieval using textual and visual features. We explore the performance bounds of a faceted system by carrying out a simulated user evaluation on TRECVid data sets, and also on the logs of a prior user experiment with the system. We first present a methodology to reduce the dimensionality of features by selecting the most important ones. Then, we discuss the simulated evaluation strategies employed in our evaluation and the effect on the use of both textual and visual features. Facets created by users are simulated by clustering video shots using textual and visual features. The experimental results of our study demonstrate that the faceted browser can potentially improve the search effectiveness.

An aspectual interface for supporting complex search tasks

With the increasing importance of search systems on the web, there is a continuing push to design interfaces which are a better match with the kinds of real-world tasks in which users are engaged. In this paper, we consider how broad, complex search tasks may be supported via the search interface. In particular, we consider search tasks which may be composed of multiple aspects, or multiple related subtasks. For example, in decision making tasks the user may investigate multiple possible solutions before settling on a single, final solution, while other tasks, such as report writing, may involve searching on multiple interrelated topics. A search interface is presented which is designed to support such broad search tasks, allowing a user to create search aspects, each of which models an independent subtask of some larger task. The interface is built on the intuition that users should be able to structure their searching environment when engaged on complex search tasks, where the act of structuring and organization may aid the user in understanding his or her task. A user study was carried out which compared our aspectual interface to a standard web-search interface. The results suggest that an aspectual interface can aid users when engaged in broad search tasks where the search aspects must be identified during searching; for a task where search aspects were pre-defined, no advantage over the baseline was found. Results for a decision making task were less clear cut, but show some evidence for improved task performance.

Comparison of Feature Construction Methods for Video Relevance Prediction

Low level features of multimedia content often have limited power to discriminate a documents relevance to a query. This motivated researchers to investigate other types of features. In this paper, we investigated four groups of features: low-level object features, behavioural features, vocabulary features, and window-based vocabulary features, to predict the relevance of shots in video retrieval. Search logs from two user studies formed the basis of our evaluation. The experimental results show that the window-based vocabulary features performed best. The behavioural features also showed a promising result, which is useful when the vocabulary features are not available. We also discuss the performance of classifiers.

A flexible news filtering model exploiting a hierarchical fuzzy categorization

In this paper we present a novel news filtering model based on flexible and soft filtering criteria and exploiting a fuzzy hierarchical categorization of news. The filtering module is designed to provide news professionals and general users with an interactive and personalised tool for news gathering and delivery. It exploits content-based filtering criteria and category-based filtering techniques to deliver to the user a ranked list of either news or clusters of news. In fact, if the user prefers to have a synthetic view of the topics of recent news pushed by the stream, the system filters groups (clusters) of news having homogenous contents, identified automatically by the application of a fuzzy clustering algorithm that organizes the recent news into a fuzzy hierarchy. The filter can be trained explicitly by the user to learn his/her interests as well as implicitly by monitoring his/her interaction with the system. Several filtering criteria can be applied to select and rank news to the users based on the users information preferences and presentation preferences. User preferences specify what information (the contents of interest) is relevant to the user, the sources that provide reliable information, and the period of time during which the information remains relevant. Each individual news or cluster of news homogeneous with respect to their content is selected based on a customizable multi criteria decision making approach and ranked based on a combination of criteria specified by the user in his/her presentation preferences.

Can an intermediary collection help users search image databases without annotations

Developing methods for searching image databases is a challenging and ongoing area of research. A common approach is to use manual annotations, although generating annotations can be expensive in terms of time and money, and therefore may not be justified in many situations. Content-based search techniques which extract visual features from image data can be used, but users are typically forced to express their information need using example images, or through sketching interfaces. This can be difficult if no visual example of the information need is available, or when the information need cannot be easily drawn. In this paper, we consider an alternative approach which allows a user to search for images through an intermediate database. In this approach, a user can search using text in the intermediate database as a way of finding visual examples of their information need. The visual examples can then be used to search a database that lacks annotations. Three experiments are presented which investigate this process. The first experiment automatically selects the image queries from the intermediary database; the second instead uses images which have been hand-picked by users. A third experiment, an interactive study, is then presented this study compares the intermediary interface to text search, where we consider text as an upper bound of performance. For this last study, an interface which supports the intermediary search process is described. Results show that while performance does not match manual annotations, users are able to find relevant material without requiring collection annotations.