Search (45 results, page 3 of 3)

0.0068065543 = product of:
  0.013613109 = sum of:
    0.013613109 = product of:
      0.027226217 = sum of:
        0.027226217 = weight(_text_:systems in 2836) [ClassicSimilarity], result of:
          0.027226217 = score(doc=2836,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1697705 = fieldWeight in 2836, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0390625 = fieldNorm(doc=2836)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Classification is the process of assigning objects to pre-defined classes based on observations or characteristics of those objects, and there are many approaches to performing this task. The overall objective of this study is to demonstrate the use of two learning techniques to analyze the results of a manual classification system. Our sample consisted of 1,026 documents, from the ACM Computing Classification System, classified by their authors as belonging to one of the groups of the classification system: "H.3 Information Storage and Retrieval." A singular value decomposition of the documents' weighted term-frequency matrix was used to represent each document in a 50-dimensional vector space. The analysis of the representation using both supervised (decision tree) and unsupervised (clustering) techniques suggests that two pairs of the ACM classes are closely related to each other in the vector space. Class 1 (Content Analysis and Indexing) is closely related to Class 3 (Information Search and Retrieval), and Class 4 (Systems and Software) is closely related to Class 5 (Online Information Services). Further analysis was performed to test the diffusion of the words in the two classes using both cosine and Euclidean distance.

0.0068065543 = product of:
  0.013613109 = sum of:
    0.013613109 = product of:
      0.027226217 = sum of:
        0.027226217 = weight(_text_:systems in 3311) [ClassicSimilarity], result of:
          0.027226217 = score(doc=3311,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1697705 = fieldWeight in 3311, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0390625 = fieldNorm(doc=3311)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Tools for automatic subject assignment help deal with scale and sustainability in creating and enriching metadata, establishing more connections across and between resources and enhancing consistency. Although some software vendors and experimental researchers claim the tools can replace manual subject indexing, hard scientific evidence of their performance in operating information environments is scarce. A major reason for this is that research is usually conducted in laboratory conditions, excluding the complexities of real-life systems and situations. The article reviews and discusses issues with existing evaluation approaches such as problems of aboutness and relevance assessments, implying the need to use more than a single "gold standard" method when evaluating indexing and retrieval, and proposes a comprehensive evaluation framework. The framework is informed by a systematic review of the literature on evaluation approaches: evaluating indexing quality directly through assessment by an evaluator or through comparison with a gold standard, evaluating the quality of computer-assisted indexing directly in the context of an indexing workflow, and evaluating indexing quality indirectly through analyzing retrieval performance.

0.0054452433 = product of:
  0.010890487 = sum of:
    0.010890487 = product of:
      0.021780973 = sum of:
        0.021780973 = weight(_text_:systems in 4197) [ClassicSimilarity], result of:
          0.021780973 = score(doc=4197,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1358164 = fieldWeight in 4197, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.03125 = fieldNorm(doc=4197)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Das unüberschaubare und permanent wachsende Angebot von Informationen im Internet ermöglicht es den Menschen nicht mehr, dieses inhaltlich zu erfassen oder gezielt nach Informationen zu suchen. Einen Lösungsweg zur verbesserten Informationsauffindung stellt hierbei die Kategorisierung bzw. Klassifikation der Informationen auf Basis ihres thematischen Inhaltes dar. Diese thematische Klassifikation kann sowohl anhand manueller (intellektueller) Methoden als auch durch automatisierte Verfahren erfolgen. Doch beide Ansätze für sich konnten die an sie gestellten Erwartungen bis zum heutigen Tag nur unzureichend erfüllen. Im Rahmen dieser Arbeit soll daher der naheliegende Ansatz, die beiden Methoden sinnvoll zu verknüpfen, untersucht werden. Im ersten Teil dieser Arbeit, dem Untersuchungsbereich, wird einleitend das Problem des Informationsüberangebots in unserer Gesellschaft erläutert und gezeigt, dass die Kategorisierung bzw. Klassifikation dieser Informationen speziell im Internet sinnvoll erscheint. Die prinzipiellen Möglichkeiten der Themenzuordnung von Dokumenten zur Verbesserung der Wissensverwaltung und Wissensauffindung werden beschrieben. Dabei werden unter anderem verschiedene Klassifikationsschemata, Topic Maps und semantische Netze vorgestellt. Schwerpunkt des Untersuchungsbereiches ist die Beschreibung automatisierter Methoden zur Themenzuordnung. Neben einem Überblick über die gebräuchlichsten Klassifikations-Algorithmen werden sowohl am Markt existierende Systeme sowie Forschungsansätze und frei verfügbare Module zur automatischen Klassifikation vorgestellt. Berücksichtigt werden auch Systeme, die zumindest teilweise den erwähnten Ansatz der Kombination von manuellen und automatischen Methoden unterstützen. Auch die in Zusammenhang mit der Klassifikation von Dokumenten im Internet auftretenden Probleme werden aufgezeigt. Die im Untersuchungsbereich gewonnenen Erkenntnisse fließen in die Entwicklung eines Moduls zur benutzerunterstützten, automatischen Dokumentklassifikation im Rahmen des xFIND Systems (extended Framework for Information Discovery) ein. Dieses an der technischen Universität Graz konzipierte Framework stellt die Basis für eine Vielzahl neuer Ideen zur Verbesserung des Information Retrieval dar. Der im Gestaltungsbereich entwickelte Lösungsansatz sieht zunächst die Verwendung bereits im System vorhandener, manuell klassifizierter Dokumente, Server oder Serverbereiche als Grundlage für die automatische Klassifikation vor. Nach erfolgter automatischer Klassifikation können in einem nächsten Schritt dann Autoren und Administratoren die Ergebnisse im Rahmen einer Benutzerunterstützung anpassen. Dabei kann das kollektive Benutzerverhalten durch die Möglichkeit eines Votings - mittels Zustimmung bzw. Ablehnung der Klassifikationsergebnisse - Einfluss finden. Das Wissen von Fachexperten und Benutzern trägt somit letztendlich zur Verbesserung der automatischen Klassifikation bei. Im Gestaltungsbereich werden die grundlegenden Konzepte, der Aufbau und die Funktionsweise des entwickelten Moduls beschrieben, sowie eine Reihe von Vorschlägen und Ideen zur Weiterentwicklung der benutzerunterstützten automatischen Dokumentklassifikation präsentiert.

0.004764588 = product of:
  0.009529176 = sum of:
    0.009529176 = product of:
      0.019058352 = sum of:
        0.019058352 = weight(_text_:systems in 3647) [ClassicSimilarity], result of:
          0.019058352 = score(doc=3647,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.118839346 = fieldWeight in 3647, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.02734375 = fieldNorm(doc=3647)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

0.0040839324 = product of:
  0.008167865 = sum of:
    0.008167865 = product of:
      0.01633573 = sum of:
        0.01633573 = weight(_text_:systems in 1253) [ClassicSimilarity], result of:
          0.01633573 = score(doc=1253,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1018623 = fieldWeight in 1253, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0234375 = fieldNorm(doc=1253)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

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.