Catálogo de publicaciones - libros

Personalized Recommendation System is a necessary part of Personalized Information Service System. In this paper, a personalized recommendation system is implemented and tested with actual website access data. According to the test result analysis, we present some new methods to improve the recommendation effectiveness.

To date, the adoption of content-based Distributed Information Retrieval (DIR) solutions within the domain of Digital Libraries (DL) remains rather limited. We contend that the lack of open application frameworks is a major obstacle. Not only to the rapid development and reuse of state-of-the art DIR functionality, but also to the holistic evaluation of competing DIR approaches. In this paper, we present the design and implementation of one such framework; as well as its use within the context of the DILIGENT infrastructure: a testbed of dedicated hardware, middleware and application services built explicitly for the DL domain upon the lower-level facilities of the first European GRID platform (EGEE). We conclude with a summary of future plans for its large scale deployment, testing and evaluation within this context.

Today, users expect a variety of digital libraries to be searchable from a single Web page. The German Vascoda project provides this service for dozens of information sources. Its ultimate goal is to provide search quality close to the ranking of a central database containing documents from all participating libraries. Currently, however, the Vascoda portal is based on a non-cooperative metasearch approach, where results from sources are merged randomly and ranking quality is sub-optimal. In this paper, we describe a Lucene-based plugin which replaces this method by a truly federated search across different search engines, where the exchange of document statistics improves document ranking. Preliminary evaluation results show ranking results equal to a centralized setup.

Objective relevance regards retrieved documents as relevant without considering users’ tasks. Subjective relevance, in contrast, focuses on usefulness of documents for users’ contexts. This paper aims to enhance objective relevance and address its limitations by designing a mock-up interface based on ACM Digital Library so as to enhance its support for users’ evaluating the subjective relevance of documents. Important features elicited using a factor analytic approach from an earlier study were used to inform the design of the search, results list and document record pages in the mock-up. A pilot study was conducted to gather users’ feedback about usefulness of designed features. Findings indicated that majority of important features designed were useful. However, subjects suggested that the design of some features could be further improved to facilitate their support for SR. The paper concludes by discussing recommendations for improving the mock-up.

Query clustering helps users frame an optimum query to obtain relevant documents. The content-based approach to query clustering has been criticized since queries are usually very short and consist of a wide variety of keywords, making this method ineffective in finding clusters. Clustering based on similar search results URLs has also performed inadequately due to the large number of distinct URLs. Our previous work has demonstrated that a hybrid approach combining the two is effective in generating good clusters. This study aims to extend our work by using lexical knowledge from WordNet to examine the effect on the quality of query clusters. Our results show that surprisingly, the use of lexical knowledge does not produce any significant improvement in quality, thus demonstrating the robustness of the hybrid clustering approach.

This paper describes a query generation using semantic features to represent the information demand of users for question answering and information retrieval. One of fundamental reasons why unwanted results are included in responses of all information retrieval systems is because queries do not exactly represent the information demand of users. To solve this problem, a query generaton using the semantic feature is intended to extract semantic features which appear commonly in natural language questions of similar type and utilize them for question answering and information retrieval. We extract semantic features from natural language questions using a grammar and generate queries which represent enough information demands of users using semantic features and syntactic structures. For performance improvement of question answering and information retrieval, we introduce a query-document similarity used to rank documents which include generated queries in the high position. We evaluated our mechanism using 100 queries about a person in the web. There was a notable improvement in the precision at N documents when our approach is applied. Especially, we found that an efficient document retrieval is possible by a question analysis based on semantic features on natural language questions which are comparatively short but fully expressing the information demand of users.

Conventional Web search engines rank their searched results page by page. That is, conventionally, the information unit for both searching and ranking is a single Web page. There are, however, cases where a set of searched pages shows a better similarity (relevance) to a given (keyword) query than each individually searched page. This is because the information a user wishes to have is sometimes distributed on multiple Web pages. In such cases, the information unit used for ranking should be a set of pages rather than a single page. In this paper, we propose the notion of a “page set ranking”, which is to rank each pertinent set of searched Web pages. We describe our new algorithm of the page set ranking to efficiently construct and rank page sets. We present some experimental results and the effectiveness of our approach.

This paper outlines the adaptation of IFLA’s Functional Requirements for Bibliographic Records (FRBR) for development of a metadata scheme to represent palm leaf manuscripts (PLMs) and facilitate their retrieval in digital collections. The FRBR model uses a structured, four-level hierarchy to represent an intellectual work with multiple titles, editions or formats. Because FRBR focuses on representation of the conceptual work rather than the physical entity, it must be modified for representation of PLMs. In this modified model, the level of work applies to the physical PLM rather than its conceptual content; expression applies to the languages in which the PLM occurs; manifestation applies to the formats in which each expression is available; and item applies to individual copies of a single format. The modified model has been used to devise a metadata scheme where each level has its own set of elements.

UNIMARC is a family of bibliographic metadata schemas with formats for descriptive information, classification, authorities and holdings. This paper describes the problem of maintenance and management of these schemas, with their online publishing for human and machine readable. This is accomplished by a Model View Controller (MVC) architecture managing the format schemas and schema descriptions in XML, as model representations, and XSLT transformations for the publishing in HTML, PDF, and other possible option. This work is an activity of a larger project toward the development of the international UNIMARC Metadata Registry, in the scope of the IFLA-CDNL Alliance for Bibliographic Standards activities.

Resources located in digital libraries are labeled (or classified) based on taxonomies. On multiple digital libraries, however, heterogeneity between taxonomies is a serious problem for efficient interoperation processes (e.g., information sharing and query transformation). In order to overcome this problem, we propose a novel framework based on aligning taxonomies of digital libraries. Thereby, the best mapping between concepts has to be discovered to maximize the summation of a set of partial similarities. For experimentation, three digital libraries were built based on different taxonomies. Taxonomy alignment-based resource retrieval was evaluated by human experts, and we measured recall and precision measures retrieved by concept replacement strategy.