Search results

This article describes the fastest growing category of machine‐readable data‐bases — full‐text databases. A selection of articles from the literature on full‐text databases was explored and this provides a basis for the information presented here on search strategy, performance measurement, and benefits and limitations of full‐text databases. Various use studies and uses of full‐text databases have also been listed.

Describes an approach to the processing and presentation of online full‐text journals that utilizes several evolving information technologies, including extensible markup language (XML) and extensible stylesheet language transformations (XSLT). Discusses major issues and trade‐offs associated with these technologies, and also specific lessons learned from our use of these technologies in the Illinois Testbed of full‐text journal articles. Focuses especially on issues associated with the representation of documents in XML, techniques to create and normalize metadata describing XML document instances, XSLT features employed in the Illinois Testbed, and trade‐offs of different XSLT implementation options. Pays special attention to techniques for transforming between XML and HTML formats for rendering in today’s commercial Web browsers.

Full text of a document is a rich source of information that can be used to provide meaningful topics. The purpose of this paper is to demonstrate how to use citation context (CC) in the full text to identify the cited topics and citing topics efficiently and effectively by employing automatic text analysis algorithms.

The authors present two novel topic models, Citation-Context-LDA (CC-LDA) and Citation-Context-Reference-LDA (CCRef-LDA). CC is leveraged to extract the citing text from the full text, which makes it possible to discover topics with accuracy. CC-LDA incorporates CC, citing text, and their latent relationship, while CCRef-LDA incorporates CC, citing text, their latent relationship and reference information in CC. Collapsed Gibbs sampling is used to achieve an approximate estimation. The capacity of CC-LDA to simultaneously learn cited topics and citing topics together with their links is investigated. Moreover, a topic influence measure method based on CC-LDA is proposed and applied to create links between the two-level topics. In addition, the capacity of CCRef-LDA to discover topic influential references is also investigated.

The results indicate CC-LDA and CCRef-LDA achieve improved or comparable performance in terms of both perplexity and symmetric Kullback–Leibler (sKL) divergence. Moreover, CC-LDA is effective in discovering the cited topics and citing topics with topic influence, and CCRef-LDA is able to find the cited topic influential references.

The automatic method provides novel knowledge for cited topics and citing topics discovery. Topic influence learnt by our model can link two-level topics and create a semantic topic network. The method can also use topic specificity as a feature to rank references.

This paper reports the results of the information retrieval project PADOK‐II. This project, which began in November 1987, is being carried out by the Linguistic Information Science Group of the University of Regensburg (LIR) in cooperation with the German Patent Office (GPO) and is sponsored by the German Ministry for Research and Technology. The long‐term aim is to integrate artificial intelligence into information retrieval research without neglecting traditional information retrieval methodology. In PADOK‐II an information retrieval system is considered which indexes documents rather shallowly using free‐text or morphological components. A large‐scale retrieval test has been carried out, based on the German Patent Information System. Answers have been obtained to some 400 queries made by 10 users in simulated real‐life situations. These results have been used to attempt to answer the question: ‘How do the linguistically‐based functions of an indexing system contribute to its performance?’ As a spinoff of this test, the influence of document size and structure was studied with a view to identifying the most reasonable basic content for a German Patent Information System.

Informally computer scientists reported they could access free copies of research papers they needed via tools such as Google Scholar. To ascertain whether this perception came from widespread free access or from unnoticed employer-paid access, the purpose of this paper was to locate computer science papers and determine what proportion was freely available using Google Scholar.

A sample of 1,967 conference papers and periodical papers from 2003 to 2010, indexed in the ACM Guide to Computing Literature, was searched for manually in Google Scholar, using the paper or article title and the first author's surname and supplementary searches as needed.

Free full-text versions were found for 52 per cent of the conference papers and 55 per cent of the periodical papers. Documents with older publication dates were more likely to be freely accessible than newer documents, with free versions found for 71 per cent of items published in 2003 and 43 per cent of items published in 2010.

Results were limited to the retrieval of known computer science publications via Google Scholar. Future research could examine whether the decline found in this study is specific to Google Scholar or reflects a decrease in the free sharing of research by computer scientists.

Previous research for computer science found lower levels of free access than this research determined, but the decline found in this study runs contrary to increases that have been found. This research confirms many computer science papers are available for free but also that subscription holdings are necessary for complete coverage of papers in the field.

Commercial document supply and full‐text online databases play significant roles in the shift from “ownership” to “access”. The escalating costs of science and technology journals, budgetary constraints, and availability of sci‐tech literature via non‐traditional sources, such as commercial document supply and full‐text online databases, are reshaping academic libraries’ sci‐tech collections, as well as the modes of accessing and delivering scientific information. Examines the value and the effectiveness of commercial document supply services and full‐text online databases in accessing sci‐tech information and discusses the impacts of budgetary limits and technology on academic libraries acquiring sci‐tech literature.

The purpose of this paper is to discuss the issue of relevance in full‐text patent document searches from the viewpoint of end‐users in science and technology. It aims to present three cases of patent document analysis for relevance, with an additional case of improved search profile with increased relevance, and to summarise the findings in the form of instructions for users.

Two methodological approaches were used for the analysis of patent documents: value‐added processing of the bibliographic part of patent documents for the identification of trends; and structuring of data into systems for the determination of patent relevance. Overall, four sets of full‐text patent documents were analysed, covering the topics of: microencapsulated phase change materials; digital photography and image sensors; patent document processing; and patent analysis.

Value‐added analysis of the bibliographic parts of patent documents is a quick and useful option for the recognition of research trends. However, where non‐relevant patent documents are present in a data set, automatic bibliographic analysis may lead to conclusions that are mathematically and statistically correct, but that are not reliable or may even be incorrect for the user's research. Non‐adequate terminology is one of the main obstacles to relevant patent searches, especially if well‐defined keywords are non‐existent, as with cases of newly emerging and fast developing scientific and technological fields.

Based on the bibliographic and content analyses of patent documents, the paper provides instructions for users in the form of ten general rules for increasing the relevance of full‐text patent document searches.

Searchers need to develop a set of heuristics (a variety of problem solving tactics and operations) for searching full‐text databases. Those heuristics which have evolved for searching bibliographic databases are often not effective when applied to full‐text. Presented here are methods which can be helpful in increasing precision when searching full‐text journal files for items on a given topic at BRS. Mead Data Central and Dialog. All of the techniques are based on achieving some repetition of central concepts as expressed in natural language. Research in automatic indexing and in full‐text retrieval provides a theoretical justification for a heuristic of this type. Schematic representations of strategies and sample searches which represent application of the techniques on each of the three systems are provided.

A microcomputer‐based optical disk system provides the user with a capability for combined full text search and image retrieval. Conceived initially to demonstrate a potential tool to assist lawyers in adjudicatory proceedings, the system appears to have potential for addressing a broad spectrum of record management requirements.

The Commission of the European Communities initiated over 1983—1986 a set of experiments in electronic publishing and electronic document delivery, in order to test the performance, costs and user acceptance of such systems. The experiments aimed to break the mould of conventional electronic information services set by the limitations of the ASCII character set and packet‐switched data networks. Full texts, including graphics, diagrams, mathematical formulae and different typefaces were experimented upon, as regards their electronic capture, storage, retrieval and distribution. A systematic independent evaluation of the experiments was carried out. This paper gives an overview of some of the results obtained, and of the major findings or spin‐offs which have been identified, which go to show that the impact of DOCDEL upon the information market has been considerable.