A new metric for multimedia retrieval in structured documents

Lecture Notes in Computer Science, 2006

Many XML documents contain a mixture of text and images. Images play an important role in webpage or article presentation. However, popular Information Retrieval systems still largely depend on pure text retrieval as it is believed that text descriptions including body text and the caption of images contain precise information. On the other hand, images are more attractive and easier to understand than pure text. We assume that if the image content is used in addition to the pure text-based retrieval, the retrieval result should be better than text-only or image-only retrieval. We test this hypothesis by doing a series of experiments using the Lonely Planet XML document collection. Two search engines, an XML document search engine using both content and structure based on text, and a content-based image search engine were used at the same time. The results generated by these two search engines were merged together to form a new result. This paper presents our current work, initial results and vision into future work.

The processing of massive amounts of data has become indispensable especially with the potential proliferation of big data. The volume of information available nowadays makes it difficult for the user to find relevant information in a vast collection of documents. As a result, the exploitation of vast document collections necessitates the implementation of automated technologies that enable appropriate and effective retrieval. In this paper, we will examine the state of the art of IR in XML documents. We will also discuss some works that have used graphs to represent documents in the context of IR. In the same vein, the relationships between the components of a graph are the center of our attention.

Image

This paper presents a formalization of an image retrieval system based on a notion of similarity between images in a multimedia database (namely XML-Enabled Database) and where a user request can be an image file or a keyword. The CBIR (Content Based Image Retrieval) system and the current search engines (e.g. Google, Yahoo….) make image search possible only when the query is a keyword. This type of search is limited because keywords are not expressive enough to describe all important characteristics of an image. For example, an exact match request cannot be formulated in such systems. Thus, we propose a search system in which a request might be an image file or a keyword. The MPEG-7 standard is used for describing an image as an XML document. A similarity distance between images is defined which is used to compare the request image with the images of a database. We also propose an algorithm to calculate a similarity distance between two XML nodes with a given precision 'k' (k is defined by the user: he can fix 'k' at 100% for the exact match retrieval of features) so as to be able to provide accurate information in response to a user request. The statistics show that our system is more efficient than leading content based image retrieval systems such as ERIC7 and current search engines.

Lecture Notes in Computer Science, 1997

We present a model for complex documents possibly consisting of a hierarchically structured set of images or texts. Documents are represented both at the form level (as sets of physical features of the representing objects), at the content level (as sets of properties of the represented entities), and at the structure level. A uniform and powerful query language allows queries to be issued that transparently combine features pertaining to form, content and structure alike. Queries are expressions of a (fuzzy) logical language. While that part of the query that pertains to (medium-independent) content is "directly" processed by an inferential engine, that part that pertains to (medium-dependent) form is entrusted to specialised document processing procedures linked to the logical language by a procedural attachment mechanism. The model thus combines the power of state-of-the-art document processing techniques with the advantages of a clean, logically defined framework for understanding multimedia document retrieval.

By rapid development of using extensible language and XML development on the Internet, retrieval of XML data has become one of the most interesting research matters. Since the XML documents are increasingly expanding, engines for search and retrieval can be developed into a set of XML documents in order to perform the search. XML documents have not only textual information, but also contain information about the logical structure of the documents. The logical structure in fact is a tree-like structure that is encrypted by the XML labels. In XML retrieval, elements and components of document are retrieved, not the whole document. Content-based retrieval of XML documents over the past few years has been the most highly regarded which mainly has emerged from the NEXI initiative design [1]. The aim of XML retrieval is restoring related parts of an XML document that by exploiting the document structure can respond to users' needs [2]. Information retrieval systems are often inconsistent with relational databases. In XML retrieval, information needs of users determine as queries, includes key phrases and structured points. Structure, specifies XML elements tracks marked in the set from which system should restore the information [3]. In XML documents and texts, structure and content are separable . An information retrieval system in response to a query returns a ranked list of documents. Then, user examine in the linear case each of them that are in a higher rank . Since the numbers of XML components are generally high, it is necessary that users have systems to retrieve XML, so that components of content have became retrieved and reviewed. One approach could involve the use of summarization that is useful in interactive information retrieval. In interactive XML retrieval, a summary can connect by any one of its document parts which has returned via XML retrieval system [6].

2007

Determining the effectiveness of XML retrieval systems is crucial for improving information retrieval from XML document collections. Traditional effectiveness measures do not address the problem of overlap in the recall-base. At the Initiative for the Evaluation of XML retrieval (INEX), extended cumulated gain (XCG) was developed to address overlap. It works by comparing the cumulated score of a retrieval result to an ideal result. The use of XCG is contingent on being able to define an ideal recall-base for every topic.

Computer Networks and ISDN Systems, 1998

While pages on the Web contain more and more multimedia information, such as images, videos and audio, today's search engines are mostly based on textual information. There is an emerging need for a new generation of search engines that try to exploit the full multimedia information present on the Web. The approach presented in this paper is based on a multimedia model intended to describe the various multimedia components, their structure and their relationships with a pre-defined taxonomy of concepts, in order to support the information retrieval process. 0 1998 Publ i shed by Elscvier Science B.V. All rights reserved. Kqwords: Search engines: Multimedia information retrieval ' http://www.itnm.columbia.edu/webseek " http://www.ccrl.neclab.com/nmorc 0169.7552/98/$I9.00 0 I998 Publ i shed by Elsevier Science B.V. All rights re\ervcd P/I SOl69-7552(98)00096-S

2006

This paper investigates the measures of retrieval effectiveness employed in the ad-hoc track of INEX 2006. In particular, it looks at how sensitive the different metrics are to various parameters and how the choice of certain parameter values may influence the conclusions of the evaluation. One such parameter-which is the main focus of the investigation-is the methodology employed in generating an ideal recallbase for the evaluation of the Focused task. The paper also examines the correlation between the different INEX measures and traditional IR metrics. methods for the construction of the ideal recall-base are proposed and compared. More importantly, however, the paper investigates how the evaluation results are actually affected by the choice of methodology. In addition, a comparison is made between the measures used at INEX and those used traditionally in IR, e.g. precision and recall.

Lecture Notes in Computer Science, 2006

Structured document retrieval aims at exploiting the structure together with the content of documents to improve retrieval results. Several aspects of traditional information retrieval applied on flat documents have to be reconsidered. These include in particular, document representation, storage, indexing, retrieval, and ranking. This paper outlines the architecture of our system and the adaptation of the standard vector space model to achieve focussed retrieval.

2005

Two common approaches in retrieving images from a collection are retrieval by text keywords and retrieval by visual content. However, it is widely recognised that it is impossible for keywords alone to fully describe visual content. This paper reports on the participation of the RMIT University group in the INEX 2005 multimedia track, where we investigated our approach of combining evidence from a content-oriented XML retrieval system and a content-based image retrieval system using a linear combination of evidence. Our approach yielded the best overall result for the INEX 2005 Multimedia track using the standard evaluation measures. We have extended our work by varying the parameter for the linear combination of evidence, and we have also examined the performance of runs submitted by participants by using the newly proposed HiXEval evaluation metric. We show that using CBIR in conjunction with text search leads to better retrieval performance.