Search (85 results, page 3 of 5)

0.0023806747 = product of:
  0.023806747 = sum of:
    0.023806747 = weight(_text_:web in 831) [ClassicSimilarity], result of:
      0.023806747 = score(doc=831,freq=4.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.25496176 = fieldWeight in 831, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0390625 = fieldNorm(doc=831)
  0.1 = coord(1/10)

Abstract

Purpose - The purpose of this research is to compare several machine learning techniques on the task of Asian language text classification, such as Chinese and Japanese where no word boundary information is available in written text. The paper advocates a simple language modeling based approach for this task. Design/methodology/approach - Naïve Bayes, maximum entropy model, support vector machines, and language modeling approaches were implemented and were applied to Chinese and Japanese text classification. To investigate the influence of word segmentation, different word segmentation approaches were investigated and applied to Chinese text. A segmentation-based approach was compared with the non-segmentation-based approach. Findings - There were two findings: the experiments show that statistical language modeling can significantly outperform standard techniques, given the same set of features; and it was found that classification with word level features normally yields improved classification performance, but that classification performance is not monotonically related to segmentation accuracy. In particular, classification performance may initially improve with increased segmentation accuracy, but eventually classification performance stops improving, and can in fact even decrease, after a certain level of segmentation accuracy. Practical implications - Apply the findings to real web text classification is ongoing work. Originality/value - The paper is very relevant to Chinese and Japanese information processing, e.g. webpage classification, web search.

0.0023806747 = product of:
  0.023806747 = sum of:
    0.023806747 = weight(_text_:web in 3614) [ClassicSimilarity], result of:
      0.023806747 = score(doc=3614,freq=4.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.25496176 = fieldWeight in 3614, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3614)
  0.1 = coord(1/10)

Abstract

Purpose - The purpose of this study is twofold: to investigate whether it is meaningful to use the Engineering Index (Ei) classification scheme for browsing, and then, if proven useful, to investigate the performance of an automated classification algorithm based on the Ei classification scheme. Design/methodology/approach - A user study was conducted in which users solved four controlled searching tasks. The users browsed the Ei classification scheme in order to examine the suitability of the classification systems for browsing. The classification algorithm was evaluated by the users who judged the correctness of the automatically assigned classes. Findings - The study showed that the Ei classification scheme is suited for browsing. Automatically assigned classes were on average partly correct, with some classes working better than others. Success of browsing showed to be correlated and dependent on classification correctness. Research limitations/implications - Further research should address problems of disparate evaluations of one and the same web page. Additional reasons behind browsing failures in the Ei classification scheme also need further investigation. Practical implications - Improvements for browsing were identified: describing class captions and/or listing their subclasses from start; allowing for searching for words from class captions with synonym search (easily provided for Ei since the classes are mapped to thesauri terms); when searching for class captions, returning the hierarchical tree expanded around the class in which caption the search term is found. The need for improvements of classification schemes was also indicated. Originality/value - A user-based evaluation of automated subject classification in the context of browsing has not been conducted before; hence the study also presents new findings concerning methodology.

0.0023567479 = product of:
  0.023567477 = sum of:
    0.023567477 = weight(_text_:web in 7209) [ClassicSimilarity], result of:
      0.023567477 = score(doc=7209,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.25239927 = fieldWeight in 7209, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0546875 = fieldNorm(doc=7209)
  0.1 = coord(1/10)

Abstract

The Nordic WAIS/WWW project sponsored by NORDINFO is a joint project between Lund University Library and the National Technological Library of Denmark. It aims to improve the existing networked information discovery and retrieval tools Wide Area Information System (WAIS) and World Wide Web (WWW), and to move towards unifying WWW and WAIS. Details current results focusing on the WAIS side of the project. Describes research into automatic indexing and classification of WAIS sources, development of an orientation tool for WAIS, and development of a WAIS index of WWW resources

0.0023567479 = product of:
  0.023567477 = sum of:
    0.023567477 = weight(_text_:web in 1568) [ClassicSimilarity], result of:
      0.023567477 = score(doc=1568,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.25239927 = fieldWeight in 1568, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0546875 = fieldNorm(doc=1568)
  0.1 = coord(1/10)

0.0023567479 = product of:
  0.023567477 = sum of:
    0.023567477 = weight(_text_:web in 1071) [ClassicSimilarity], result of:
      0.023567477 = score(doc=1071,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.25239927 = fieldWeight in 1071, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0546875 = fieldNorm(doc=1071)
  0.1 = coord(1/10)

Abstract

A fuzzy multiset model for information clustering is proposed with application to information retrieval on the World Wide Web. Noting that a search engine retrieves multiple occurrences of the same subjects with possibly different degrees of relevance, we observe that fuzzy multisets provide an appropriate model of information retrieval on the WWW. Information clustering which means both term clustering and document clustering is considered. Three methods of the hard c-means, fuzzy c-means, and an agglomerative method using cluster centers are proposed. Two distances between fuzzy multisets and algorithms for calculating cluster centers are defined. Theoretical properties concerning the clustering algorithms are studied. Illustrative examples are given to show how the algorithms work.

0.0023325754 = product of:
  0.023325754 = sum of:
    0.023325754 = weight(_text_:web in 2487) [ClassicSimilarity], result of:
      0.023325754 = score(doc=2487,freq=6.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.24981049 = fieldWeight in 2487, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.03125 = fieldNorm(doc=2487)
  0.1 = coord(1/10)

Abstract

Automatisches Klassifizieren von Textdokumenten bedeutet die maschinelle Zuordnung jeweils einer oder mehrerer Notationen eines vorgegebenen Klassifikationssystems zu natürlich-sprachlichen Texten mithilfe eines geeigneten Algorithmus. In der vorliegenden Arbeit wird in Form einer umfassenden Literaturstudie ein aktueller Kenntnisstand zu den Ein-satzmöglichkeiten des automatischen Klassifizierens für die sachliche Erschliessung von elektronischen Dokumenten, insbesondere von Web-Ressourcen, erarbeitet. Dies betrifft zum einen den methodischen Aspekt und zum anderen die in relevanten Projekten und Anwendungen gewonnenen Erfahrungen. In methodischer Hinsicht gelten heute statistische Verfahren, die auf dem maschinellen Lernen basieren und auf der Grundlage bereits klassifizierter Beispieldokumente ein Modell - einen "Klassifikator" - erstellen, das zur Klassifizierung neuer Dokumente verwendet werden kann, als "state-of-the-art". Die vier in den 1990er Jahren an den Universitäten Lund, Wolverhampton und Oldenburg sowie bei OCLC (Dublin, OH) durchgeführten "grossen" Projekte zum automatischen Klassifizieren von Web-Ressourcen, die in dieser Arbeit ausführlich analysiert werden, arbeiteten allerdings noch mit einfacheren bzw. älteren methodischen Ansätzen. Diese Projekte bedeuten insbesondere aufgrund ihrer Verwendung etablierter bibliothekarischer Klassifikationssysteme einen wichtigen Erfahrungsgewinn, selbst wenn sie bisher nicht zu permanenten und qualitativ zufriedenstellenden Diensten für die Erschliessung elektronischer Ressourcen geführt haben. Die Analyse der weiteren einschlägigen Anwendungen und Projekte lässt erkennen, dass derzeit in den Bereichen Patent- und Mediendokumentation die aktivsten Bestrebungen bestehen, Systeme für die automatische klassifikatorische Erschliessung elektronischer Dokumente im laufenden operativen Betrieb einzusetzen. Dabei dominieren jedoch halbautomatische Systeme, die menschliche Bearbeiter durch Klassifizierungsvorschläge unterstützen, da die gegenwärtig erreichbare Klassifizierungsgüte für eine Vollautomatisierung meist noch nicht ausreicht. Weitere interessante Anwendungen und Projekte finden sich im Bereich von Web-Portalen, Suchmaschinen und (kommerziellen) Informationsdiensten, während sich etwa im Bibliothekswesen kaum nennenswertes Interesse an einer automatischen Klassifizierung von Büchern bzw. bibliographischen Datensätzen registrieren lässt. Die Studie schliesst mit einer Diskussion der wichtigsten Projekte und Anwendungen sowie einiger im Zusammenhang mit dem automatischen Klassifizieren relevanter Fragestellungen und Themen.

0.002086211 = product of:
  0.02086211 = sum of:
    0.02086211 = product of:
      0.06258633 = sum of:
        0.06258633 = weight(_text_:29 in 5169) [ClassicSimilarity], result of:
          0.06258633 = score(doc=5169,freq=2.0), product of:
            0.10064617 = queryWeight, product of:
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.028611459 = queryNorm
            0.6218451 = fieldWeight in 5169, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.125 = fieldNorm(doc=5169)
      0.33333334 = coord(1/3)
  0.1 = coord(1/10)

Source

Nachrichten für Dokumentation. 29(1978), S.92-96

0.0020200694 = product of:
  0.020200694 = sum of:
    0.020200694 = weight(_text_:web in 316) [ClassicSimilarity], result of:
      0.020200694 = score(doc=316,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.21634221 = fieldWeight in 316, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.046875 = fieldNorm(doc=316)
  0.1 = coord(1/10)

Abstract

Information retrieval over the Internet increasingly requires the filtering of thousands of heterogeneous information sources. Important sources of information include not only traditional databases with structured data and queries, but also increasing numbers of non-traditional, semi- or unstructured collections such as Web sites, FTP archives, etc. As the number and variability of sources increases, new ways of automatically summarizing, discovering, and selecting collections relevant to a user's query are needed. One such method involves the use of classification schemes, such as the Library of Congress Classification (LCC) [10], within which a collection may be represented based on its content, irrespective of the structure of the actual data or documents. For such a system to be useful in a large-scale distributed environment, it must be easy to use for both collection managers and users. As a result, it must be possible to classify documents automatically within a classification scheme. Furthermore, there must be a straightforward and intuitive interface with which the user may use the scheme to assist in information retrieval (IR).

0.0020200694 = product of:
  0.020200694 = sum of:
    0.020200694 = weight(_text_:web in 995) [ClassicSimilarity], result of:
      0.020200694 = score(doc=995,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.21634221 = fieldWeight in 995, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.046875 = fieldNorm(doc=995)
  0.1 = coord(1/10)

Abstract

Recently, a new community has started to emerge around the development of new information research methods for searching and analyzing semi-structured and XML like documents. The goal is to handle both content and structural information, and to deal with different types of information content (text, image, etc.). We consider here the task of structured document classification. We propose a generative model able to handle both structure and content which is based on Bayesian networks. We then show how to transform this generative model into a discriminant classifier using the method of Fisher kernel. The model is then extended for dealing with different types of content information (here text and images). The model was tested on three databases: the classical webKB corpus composed of HTML pages, the new INEX corpus which has become a reference in the field of ad-hoc retrieval for XML documents, and a multimedia corpus of Web pages.

0.0020200694 = product of:
  0.020200694 = sum of:
    0.020200694 = weight(_text_:web in 1168) [ClassicSimilarity], result of:
      0.020200694 = score(doc=1168,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.21634221 = fieldWeight in 1168, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.046875 = fieldNorm(doc=1168)
  0.1 = coord(1/10)

Abstract

The Web puzzle of online information resources often hinders end-users from effective and efficient access to these resources. Clustering resources into appropriate subject-based groupings may help alleviate these difficulties, but will it work with heterogeneous material? The University of Michigan and the University of California Irvine joined forces to test automatically enhancing metadata records using the Topic Modeling algorithm on the varied OAIster corpus. We created labels for the resulting clusters of metadata records, matched the clusters to an in-house classification system, and developed a prototype that would showcase methods for search and retrieval using the enhanced records. Results indicated that while the algorithm was somewhat time-intensive to run and using a local classification scheme had its drawbacks, precise clustering of records was achieved and the prototype interface proved that faceted classification could be powerful in helping end-users find resources.

0.0016833913 = product of:
  0.016833913 = sum of:
    0.016833913 = weight(_text_:web in 1382) [ClassicSimilarity], result of:
      0.016833913 = score(doc=1382,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.18028519 = fieldWeight in 1382, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0390625 = fieldNorm(doc=1382)
  0.1 = coord(1/10)

Abstract

The authors describe their research which improves software reuse by using an automated approach to semantically search for and retrieve reusable software components in large software component repositories and on the World Wide Web (WWW). Using automation and smart (semantic) techniques, their approach speeds up the search and retrieval of reusable software components, while retaining good accuracy, and therefore improves the affordability of software reuse. A program understanding of software components and natural language understanding of user queries was employed. Then the software component descriptions were compared by matching the resulting semantic representations of the user queries to the semantic representations of the software components to search for software components that best match the user queries. A proof of concept system was developed to test the authors' approach. The results of this proof of concept system were compared to human experts, and statistical analysis was performed on the collected experimental data. The results from these experiments demonstrate that this automated semantic-based approach for software reusable component classification and retrieval is successful when compared to the labor-intensive results from the experts, thus showing that this approach can significantly benefit software reuse classification and retrieval.

0.0016833913 = product of:
  0.016833913 = sum of:
    0.016833913 = weight(_text_:web in 1056) [ClassicSimilarity], result of:
      0.016833913 = score(doc=1056,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.18028519 = fieldWeight in 1056, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0390625 = fieldNorm(doc=1056)
  0.1 = coord(1/10)

Abstract

The ever increasing amount of digitally available information is curse and blessing at the same time. On the one hand, users have increasingly large amounts of information at their fingertips. On the other hand, the assessment and refinement of web search results becomes more and more tiresome and difficult for non-experts in a domain. Therefore, established digital libraries offer specialized collections with a certain degree of quality. This quality can largely be attributed to the great effort invested into semantic enrichment of the provided documents e.g. by annotating their documents with respect to a domain-specific taxonomy. This process is still done manually in many domains, e.g. chemistry CAS, medicine MeSH, or mathematics MSC. But due to the growing amount of data, this manual task gets more and more time consuming and expensive. The only solution for this problem seems to employ automated classification algorithms, but from evaluations done in previous research, conclusions to a real world scenario are difficult to make. We therefore conducted a large scale feasibility study on a real world data set from one of the biggest mathematical digital libraries, i.e. Zentralblatt MATH, with special focus on its practical applicability.

0.0016833913 = product of:
  0.016833913 = sum of:
    0.016833913 = weight(_text_:web in 2722) [ClassicSimilarity], result of:
      0.016833913 = score(doc=2722,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.18028519 = fieldWeight in 2722, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2722)
  0.1 = coord(1/10)

Abstract

We present a Life-Long Learning from Mistakes (3LM) algorithm for document classification, which could be used in various scenarios such as spam filtering, blog classification, and web resource categorization. We extend the ideas of online clustering and batch-mode centroid-based classification to online learning with negative feedback. The 3LM is a competitive learning algorithm, which avoids over-smoothing, characteristic of the centroid-based classifiers, by using a different class representative, which we call clusterhead. The clusterheads competing for vector-space dominance are drawn toward misclassified documents, eventually bringing the model to a "balanced state" for a fixed distribution of documents. Subsequently, the clusterheads oscillate between the misclassified documents, heuristically minimizing the rate of misclassifications, an NP-complete problem. Further, the 3LM algorithm prevents over-fitting by "leashing" the clusterheads to their respective centroids. A clusterhead provably converges if its class can be separated by a hyper-plane from all other classes. Lifelong learning with fixed learning rate allows 3LM to adapt to possibly changing distribution of the data and continually learn and unlearn document classes. We report on our experiments, which demonstrate high accuracy of document classification on Reuters21578, OHSUMED, and TREC07p-spam datasets. The 3LM algorithm did not show over-fitting, while consistently outperforming centroid-based, Naïve Bayes, C4.5, AdaBoost, kNN, and SVM whose accuracy had been reported on the same three corpora.

0.0016833913 = product of:
  0.016833913 = sum of:
    0.016833913 = weight(_text_:web in 3627) [ClassicSimilarity], result of:
      0.016833913 = score(doc=3627,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.18028519 = fieldWeight in 3627, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3627)
  0.1 = coord(1/10)

Abstract

A very important extension of the traditional domain of knowledge organization (KO) arises from attempts to incorporate techniques devised in the computer science domain for automatic concept extraction and for grouping, categorizing, clustering and otherwise organizing knowledge using mechanical means. Four specific terms have emerged to identify the most prevalent techniques: machine learning, clustering, automatic indexing, and automatic classification. Our study presents three domain analytical case analyses in search of answers. The first case relies on citations located using the ISKO-supported "Knowledge Organization Bibliography." The second case relies on works in both Web of Science and SCOPUS. Case three applies co-word analysis and citation analysis to the contents of the papers in the present special issue. We observe scholars involved in "clustering" and "automatic classification" who share common thematic emphases. But we have found no coherence, no common activity and no social semantics. We have not found a research front, or a common teleology within the KO domain. We also have found a lively group of authors who have succeeded in submitting papers to this special issue, and their work quite interestingly aligns with the case studies we report. There is an emphasis on KO for information retrieval; there is much work on clustering (which involves conceptual points within texts) and automatic classification (which involves semantic groupings at the meta-document level).

0.001550583 = product of:
  0.015505831 = sum of:
    0.015505831 = product of:
      0.04651749 = sum of:
        0.04651749 = weight(_text_:22 in 1046) [ClassicSimilarity], result of:
          0.04651749 = score(doc=1046,freq=2.0), product of:
            0.10019246 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.028611459 = queryNorm
            0.46428138 = fieldWeight in 1046, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.09375 = fieldNorm(doc=1046)
      0.33333334 = coord(1/3)
  0.1 = coord(1/10)

Date

5. 5.2003 14:17:22

0.0014578596 = product of:
  0.014578596 = sum of:
    0.014578596 = weight(_text_:web in 38) [ClassicSimilarity], result of:
      0.014578596 = score(doc=38,freq=6.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.15613155 = fieldWeight in 38, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.01953125 = fieldNorm(doc=38)
  0.1 = coord(1/10)

Abstract

Automatisches Klassifizieren von Textdokumenten bedeutet die maschinelle Zuordnung jeweils einer oder mehrerer Notationen eines vorgegebenen Klassifikationssystems zu natürlich-sprachlichen Texten mithilfe eines geeigneten Algorithmus. In der vorliegenden Arbeit wird in Form einer umfassenden Literaturstudie ein aktueller Kenntnisstand zu den Ein-satzmöglichkeiten des automatischen Klassifizierens für die sachliche Erschliessung von elektronischen Dokumenten, insbesondere von Web-Ressourcen, erarbeitet. Dies betrifft zum einen den methodischen Aspekt und zum anderen die in relevanten Projekten und Anwendungen gewonnenen Erfahrungen. In methodischer Hinsicht gelten heute statistische Verfahren, die auf dem maschinellen Lernen basieren und auf der Grundlage bereits klassifizierter Beispieldokumente ein Modell - einen "Klassifikator" - erstellen, das zur Klassifizierung neuer Dokumente verwendet werden kann, als "state-of-the-art". Die vier in den 1990er Jahren an den Universitäten Lund, Wolverhampton und Oldenburg sowie bei OCLC (Dublin, OH) durchgeführten "grossen" Projekte zum automatischen Klassifizieren von Web-Ressourcen, die in dieser Arbeit ausführlich analysiert werden, arbeiteten allerdings noch mit einfacheren bzw. älteren methodischen Ansätzen. Diese Projekte bedeuten insbesondere aufgrund ihrer Verwendung etablierter bibliothekarischer Klassifikationssysteme einen wichtigen Erfahrungsgewinn, selbst wenn sie bisher nicht zu permanenten und qualitativ zufriedenstellenden Diensten für die Erschliessung elektronischer Ressourcen geführt haben. Die Analyse der weiteren einschlägigen Anwendungen und Projekte lässt erkennen, dass derzeit in den Bereichen Patent- und Mediendokumentation die aktivsten Bestrebungen bestehen, Systeme für die automatische klassifikatorische Erschliessung elektronischer Dokumente im laufenden operativen Betrieb einzusetzen. Dabei dominieren jedoch halbautomatische Systeme, die menschliche Bearbeiter durch Klassifizierungsvorschläge unterstützen, da die gegenwärtig erreichbare Klassifizierungsgüte für eine Vollautomatisierung meist noch nicht ausreicht. Weitere interessante Anwendungen und Projekte finden sich im Bereich von Web-Portalen, Suchmaschinen und (kommerziellen) Informationsdiensten, während sich etwa im Bibliothekswesen kaum nennenswertes Interesse an einer automatischen Klassifizierung von Büchern bzw. bibliographischen Datensätzen registrieren lässt. Die Studie schliesst mit einer Diskussion der wichtigsten Projekte und Anwendungen sowie einiger im Zusammenhang mit dem automatischen Klassifizieren relevanter Fragestellungen und Themen.

0.0014284048 = product of:
  0.014284048 = sum of:
    0.014284048 = weight(_text_:web in 1253) [ClassicSimilarity], result of:
      0.014284048 = score(doc=1253,freq=4.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.15297705 = fieldWeight in 1253, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.0234375 = fieldNorm(doc=1253)
  0.1 = coord(1/10)

Abstract

Information retrieval over the Internet increasingly requires the filtering of thousands of heterogeneous information sources. Important sources of information include not only traditional databases with structured data and queries, but also increasing numbers of non-traditional, semi- or unstructured collections such as Web sites, FTP archives, etc. As the number and variability of sources increases, new ways of automatically summarizing, discovering, and selecting collections relevant to a user's query are needed. One such method involves the use of classification schemes, such as the Library of Congress Classification (LCC), within which a collection may be represented based on its content, irrespective of the structure of the actual data or documents. For such a system to be useful in a large-scale distributed environment, it must be easy to use for both collection managers and users. As a result, it must be possible to classify documents automatically within a classification scheme. Furthermore, there must be a straightforward and intuitive interface with which the user may use the scheme to assist in information retrieval (IR). Our work with the Alexandria Digital Library (ADL) Project focuses on geo-referenced information, whether text, maps, aerial photographs, or satellite images. As a result, we have emphasized techniques which work with both text and non-text, such as combined textual and graphical queries, multi-dimensional indexing, and IR methods which are not solely dependent on words or phrases. Part of this work involves locating relevant online sources of information. In particular, we have designed and are currently testing aspects of an architecture, Pharos, which we believe will scale up to 1.000.000 heterogeneous sources. Pharos accommodates heterogeneity in content and format, both among multiple sources as well as within a single source. That is, we consider sources to include Web sites, FTP archives, newsgroups, and full digital libraries; all of these systems can include a wide variety of content and multimedia data formats. Pharos is based on the use of hierarchical classification schemes. These include not only well-known 'subject' (or 'concept') based schemes such as the Dewey Decimal System and the LCC, but also, for example, geographic classifications, which might be constructed as layers of smaller and smaller hierarchical longitude/latitude boxes. Pharos is designed to work with sophisticated queries which utilize subjects, geographical locations, temporal specifications, and other types of information domains. The Pharos architecture requires that hierarchically structured collection metadata be extracted so that it can be partitioned in such a way as to greatly enhance scalability. Automated classification is important to Pharos because it allows information sources to extract the requisite collection metadata automatically that must be distributed.

0.0013467129 = product of:
  0.013467129 = sum of:
    0.013467129 = weight(_text_:web in 2596) [ClassicSimilarity], result of:
      0.013467129 = score(doc=2596,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.14422815 = fieldWeight in 2596, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.03125 = fieldNorm(doc=2596)
  0.1 = coord(1/10)

Content

Ramana Rao (Inxight, Palo Alto, CA) 7 ± 2 Insights on achieving Effective Information Access Session One: Updates and a twelve month perspective Danny Sullivan (Search Engine Watch, US / England) Portalization and other search trends Carol Tenopir (University of Tennessee) Search realities faced by end users and professional searchers Session Two: Today's search engines and beyond Daniel Hoogterp (Retrieval Technologies, McLean, VA) Effective presentation and utilization of search techniques Rick Kenny (Fulcrum Technologies, Ontario, Canada) Beyond document clustering: The knowledge impact statement Gary Stock (Ingenius, Kalamazoo, MI) Automated change monitoring Gary Culliss (Direct Hit, Wellesley Hills, MA) User popularity ranked search engines Byron Dom (IBM, CA) Automatically finding the best pages on the World Wide Web (CLEVER) Peter Tomassi (LookSmart, San Francisco, CA) Adding human intellect to search technology Session Three: Panel discussion: Human v automated categorization and editing Ev Brenner (New York, NY)- Chairman James Callan (University of Massachusetts, MA) Marc Krellenstein (Northern Light Technology, Cambridge, MA) Dan Miller (Ask Jeeves, Berkeley, CA) Session Four: Updates and a twelve month perspective Steve Arnold (AIT, Harrods Creek, KY) Review: The leading edge in search and retrieval software Ellen Voorhees (NIST, Gaithersburg, MD) TREC update Session Five: Search engines now and beyond Intelligent Agents John Snyder (Muscat, Cambridge, England) Practical issues behind intelligent agents Text summarization Therese Firmin, (Dept of Defense, Ft George G. Meade, MD) The TIPSTER/SUMMAC evaluation of automatic text summarization systems Cross language searching Elizabeth Liddy (TextWise, Syracuse, NY) A conceptual interlingua approach to cross-language retrieval. Video search and retrieval Armon Amir (IBM, Almaden, CA) CueVideo: Modular system for automatic indexing and browsing of video/audio Speech recognition Michael Witbrock (Lycos, Waltham, MA) Retrieval of spoken documents Visualization James A. Wise (Integral Visuals, Richland, WA) Information visualization in the new millennium: Emerging science or passing fashion? Text mining David Evans (Claritech, Pittsburgh, PA) Text mining - towards decision support

0.0013467129 = product of:
  0.013467129 = sum of:
    0.013467129 = weight(_text_:web in 4197) [ClassicSimilarity], result of:
      0.013467129 = score(doc=4197,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.14422815 = fieldWeight in 4197, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.03125 = fieldNorm(doc=4197)
  0.1 = coord(1/10)

0.0013467129 = product of:
  0.013467129 = sum of:
    0.013467129 = weight(_text_:web in 4051) [ClassicSimilarity], result of:
      0.013467129 = score(doc=4051,freq=2.0), product of:
        0.0933738 = queryWeight, product of:
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.028611459 = queryNorm
        0.14422815 = fieldWeight in 4051, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.2635105 = idf(docFreq=4597, maxDocs=44218)
          0.03125 = fieldNorm(doc=4051)
  0.1 = coord(1/10)

Abstract

Die zunehmende Verfügbarmachung digitaler Informationen in den letzten Jahren sowie die Aussicht auf ein weiteres Ansteigen der sogenannten Datenflut kumulieren in einem grundlegenden, sich weiter verstärkenden Informationsstrukturierungsproblem. Die stetige Zunahme von digitalen Informationsressourcen im World Wide Web sichert zwar jederzeit und ortsungebunden den Zugriff auf verschiedene Informationen; offen bleibt der strukturierte Zugang, insbesondere zu wissenschaftlichen Ressourcen. Angesichts der steigenden Anzahl elektronischer Inhalte und vor dem Hintergrund stagnierender bzw. knapper werdender personeller Ressourcen in der Sacherschließun schafft keine Bibliothek bzw. kein Bibliotheksverbund es mehr, weder aktuell noch zukünftig, alle digitalen Daten zu erfassen, zu strukturieren und zueinander in Beziehung zu setzen. In der Informationsgesellschaft des 21. Jahrhunderts wird es aber zunehmend wichtiger, die in der Flut verschwundenen wissenschaftlichen Informationen zeitnah, angemessen und vollständig zu strukturieren und somit als Basis für eine Wissensgenerierung wieder nutzbar zu machen. Eine normierte Inhaltserschließung digitaler Informationsressourcen ist deshalb für die Deutsche Zentralbibliothek für Wirtschaftswissenschaften (ZBW) als wichtige Informationsinfrastruktureinrichtung in diesem Bereich ein entscheidender und auch erfolgskritischer Aspekt im Wettbewerb mit anderen Informationsdienstleistern. Weil die traditionelle intellektuelle Sacherschließung aber nicht beliebig skalierbar ist - mit dem Anstieg der Zahl an Online-Dokumenten steigt proportional auch der personelle Ressourcenbedarf an Fachreferenten, wenn ein gewisser Qualitätsstandard gehalten werden soll - bedarf es zukünftig anderer Sacherschließungsverfahren. Automatisierte Verschlagwortungsmethoden werden dabei als einzige Möglichkeit angesehen, die bibliothekarische Sacherschließung auch im digitalen Zeitalter zukunftsfest auszugestalten. Zudem können maschinelle Ansätze dazu beitragen, die Heterogenitäten (Indexierungsinkonsistenzen) zwischen den einzelnen Sacherschließer zu nivellieren, und somit zu einer homogeneren Erschließung des Bibliotheksbestandes beitragen.