Search (7 results, page 1 of 1)

0.015380534 = product of:
  0.046141602 = sum of:
    0.014799163 = weight(_text_:in in 2380) [ClassicSimilarity], result of:
      0.014799163 = score(doc=2380,freq=22.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.24922498 = fieldWeight in 2380, product of:
          4.690416 = tf(freq=22.0), with freq of:
            22.0 = termFreq=22.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2380)
    0.03134244 = weight(_text_:und in 2380) [ClassicSimilarity], result of:
      0.03134244 = score(doc=2380,freq=14.0), product of:
        0.09675359 = queryWeight, product of:
          2.216367 = idf(docFreq=13101, maxDocs=44218)
          0.043654136 = queryNorm
        0.32394084 = fieldWeight in 2380, product of:
          3.7416575 = tf(freq=14.0), with freq of:
            14.0 = termFreq=14.0
          2.216367 = idf(docFreq=13101, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2380)
  0.33333334 = coord(2/6)

Abstract

Das Internet birgt schier endlose Informationen. Ein zentrales Problem besteht heutzutage darin diese auch zugänglich zu machen. Es ist ein fundamentales Domänenwissen erforderlich, um in einer Volltextsuche die korrekten Suchanfragen zu formulieren. Das ist jedoch oftmals nicht vorhanden, so dass viel Zeit aufgewandt werden muss, um einen Überblick des behandelten Themas zu erhalten. In solchen Situationen findet sich ein Nutzer in einem explorativen Suchvorgang, in dem er sich schrittweise an ein Thema heranarbeiten muss. Für die Organisation von Daten werden mittlerweile ganz selbstverständlich Verfahren des Machine Learnings verwendet. In den meisten Fällen bleiben sie allerdings für den Anwender unsichtbar. Die interaktive Verwendung in explorativen Suchprozessen könnte die menschliche Urteilskraft enger mit der maschinellen Verarbeitung großer Datenmengen verbinden. Topic Models sind ebensolche Verfahren. Sie finden in einem Textkorpus verborgene Themen, die sich relativ gut von Menschen interpretieren lassen und sind daher vielversprechend für die Anwendung in explorativen Suchprozessen. Nutzer können damit beim Verstehen unbekannter Quellen unterstützt werden. Bei der Betrachtung entsprechender Forschungsarbeiten fiel auf, dass Topic Models vorwiegend zur Erzeugung statischer Visualisierungen verwendet werden. Das Sensemaking ist ein wesentlicher Bestandteil der explorativen Suche und wird dennoch nur in sehr geringem Umfang genutzt, um algorithmische Neuerungen zu begründen und in einen umfassenden Kontext zu setzen. Daraus leitet sich die Vermutung ab, dass die Verwendung von Modellen des Sensemakings und die nutzerzentrierte Konzeption von explorativen Suchen, neue Funktionen für die Interaktion mit Topic Models hervorbringen und einen Kontext für entsprechende Forschungsarbeiten bieten können.

Footnote

Masterthesis zur Erlangung des akademischen Grades Master of Science (M.Sc.) vorgelegt an der Fachhochschule Köln / Fakultät für Informatik und Ingenieurswissenschaften im Studiengang Medieninformatik.

Imprint

Gummersbach : Fakultät für Informatik und Ingenieurswissenschaften

Theme

Semantisches Umfeld in Indexierung u. Retrieval

0.015172194 = product of:
  0.04551658 = sum of:
    0.009977593 = weight(_text_:in in 3194) [ClassicSimilarity], result of:
      0.009977593 = score(doc=3194,freq=10.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.16802745 = fieldWeight in 3194, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3194)
    0.035538986 = weight(_text_:und in 3194) [ClassicSimilarity], result of:
      0.035538986 = score(doc=3194,freq=18.0), product of:
        0.09675359 = queryWeight, product of:
          2.216367 = idf(docFreq=13101, maxDocs=44218)
          0.043654136 = queryNorm
        0.3673144 = fieldWeight in 3194, product of:
          4.2426405 = tf(freq=18.0), with freq of:
            18.0 = termFreq=18.0
          2.216367 = idf(docFreq=13101, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3194)
  0.33333334 = coord(2/6)

Abstract

In dieser Untersuchung wurden zwei Systeme eingesetzt, um MWT aus einer Dokumentkollektion mit fachsprachlichem Bezug (Volltexte des ACL Anthology Reference Corpus) automatisch zu extrahieren. Das thematische Spektrum umfasste alle Bereiche der natürlichen Sprachverarbeitung, im Speziellen die CL als interdisziplinäre Wissenschaft. Ziel war es MWT zu extrahieren, die als potentielle Indexterme im IR Verwendung finden können. Diese sollten auf Konzepte, Methoden, Verfahren und Algorithmen in der CL und angrenzenden Teilgebieten, wie Linguistik und Informatik hinweisen bzw. benennen.
Als Extraktionssysteme wurden der TreeTagger und die Indexierungssoftware Lingo verwendet. Der TreeTagger basiert auf einem statistischen Tagging- und Chunking- Algorithmus, mit dessen Hilfe NPs automatisch identifiziert und extrahiert werden. Er kann für verschiedene Anwendungsszenarien der natürlichen Sprachverarbeitung eingesetzt werden, in erster Linie als POS-Tagger für unterschiedliche Sprachen. Das Indexierungssystem Lingo arbeitet im Gegensatz zum TreeTagger mit elektronischen Wörterbüchern und einem musterbasierten Abgleich. Lingo ist ein auf automatische Indexierung ausgerichtetes System, was eine Vielzahl von Modulen mitliefert, die individuell auf eine bestimmte Aufgabenstellung angepasst und aufeinander abgestimmt werden können. Die unterschiedlichen Verarbeitungsweisen haben sich in den Ergebnismengen beider Systeme deutlich gezeigt. Die gering ausfallenden Übereinstimmungen der Ergebnismengen verdeutlichen die abweichende Funktionsweise und konnte mit einer qualitativen Analyse beispielhaft beschrieben werden. In der vorliegenden Arbeit kann abschließend nicht geklärt werden, welches der beiden Systeme bevorzugt für die Generierung von Indextermen eingesetzt werden sollte.

0.012001218 = product of:
  0.036003653 = sum of:
    0.0075724614 = weight(_text_:in in 2849) [ClassicSimilarity], result of:
      0.0075724614 = score(doc=2849,freq=4.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.12752387 = fieldWeight in 2849, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=2849)
    0.02843119 = weight(_text_:und in 2849) [ClassicSimilarity], result of:
      0.02843119 = score(doc=2849,freq=8.0), product of:
        0.09675359 = queryWeight, product of:
          2.216367 = idf(docFreq=13101, maxDocs=44218)
          0.043654136 = queryNorm
        0.29385152 = fieldWeight in 2849, product of:
          2.828427 = tf(freq=8.0), with freq of:
            8.0 = termFreq=8.0
          2.216367 = idf(docFreq=13101, maxDocs=44218)
          0.046875 = fieldNorm(doc=2849)
  0.33333334 = coord(2/6)

Abstract

Natural Language Classifier helfen Unternehmen zunehmend dabei die Flut von Textdaten zu überwinden. Aber diese Classifier, einmal trainiert, verlieren mit der Zeit ihre Nützlichkeit. Sie bleiben statisch, aber die zugrundeliegende Domäne der Textdaten verändert sich: Ihre Genauigkeit nimmt aufgrund eines Phänomens ab, das als Konzeptdrift bekannt ist. Die Frage ist ob Konzeptdrift durch die Ausgabe eines Classifiers zuverlässig erkannt werden kann, und falls ja: ist es möglich dem durch nachtrainieren des Classifiers entgegenzuwirken. Es wird eine System-Implementierung mittels Proof-of-Concept vorgestellt, bei der das Konfidenzmass des Classifiers zur Erkennung von Konzeptdrift verwendet wird. Der Classifier wird dann iterativ neu trainiert, indem er Stichproben mit niedrigem Konfidenzmass auswählt, sie korrigiert und im Trainingsset der nächsten Iteration verwendet. Die Leistung des Classifiers wird über die Zeit gemessen, und die Leistung des Systems beobachtet. Basierend darauf werden schließlich Empfehlungen gegeben, die sich bei der Implementierung solcher Systeme als nützlich erweisen können.

Content

Diese Publikation entstand im Rahmen einer Thesis zum Master of Science FHO in Business Administration, Major Information and Data Management.

Imprint

Chur : Hochschule für Technik und Wirtschaft / Arbeitsbereich Informationswissenschaft

0.010872297 = product of:
  0.03261689 = sum of:
    0.008924231 = weight(_text_:in in 1810) [ClassicSimilarity], result of:
      0.008924231 = score(doc=1810,freq=2.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.15028831 = fieldWeight in 1810, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.078125 = fieldNorm(doc=1810)
    0.02369266 = weight(_text_:und in 1810) [ClassicSimilarity], result of:
      0.02369266 = score(doc=1810,freq=2.0), product of:
        0.09675359 = queryWeight, product of:
          2.216367 = idf(docFreq=13101, maxDocs=44218)
          0.043654136 = queryNorm
        0.24487628 = fieldWeight in 1810, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          2.216367 = idf(docFreq=13101, maxDocs=44218)
          0.078125 = fieldNorm(doc=1810)
  0.33333334 = coord(2/6)

Content

Trägerin des VFI-Dissertationspreises 2014: "Überzeugende gründliche linguistische und quantitative Analyse eines im Information Retrieval bisher wenig beachteten Textelementes anhand eines eigens erstellten grossen Hypertextkorpus, einschliesslich der Evaluation selbsterstellter Auflösungsregeln für die Nutzung in künftigen IR-Systemen.".

0.00990557 = product of:
  0.02971671 = sum of:
    0.011973113 = weight(_text_:in in 563) [ClassicSimilarity], result of:
      0.011973113 = score(doc=563,freq=10.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.20163295 = fieldWeight in 563, product of:
          3.1622777 = tf(freq=10.0), with freq of:
            10.0 = termFreq=10.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.046875 = fieldNorm(doc=563)
    0.017743597 = product of:
      0.035487194 = sum of:
        0.035487194 = weight(_text_:22 in 563) [ClassicSimilarity], result of:
          0.035487194 = score(doc=563,freq=2.0), product of:
            0.15286934 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.043654136 = queryNorm
            0.23214069 = fieldWeight in 563, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046875 = fieldNorm(doc=563)
      0.5 = coord(1/2)
  0.33333334 = coord(2/6)

Abstract

In this thesis we propose three new word association measures for multi-word term extraction. We combine these association measures with LocalMaxs algorithm in our extraction model and compare the results of different multi-word term extraction methods. Our approach is language and domain independent and requires no training data. It can be applied to such tasks as text summarization, information retrieval, and document classification. We further explore the potential of using multi-word terms as an effective representation for general web-page summarization. We extract multi-word terms from human written summaries in a large collection of web-pages, and generate the summaries by aligning document words with these multi-word terms. Our system applies machine translation technology to learn the aligning process from a training set and focuses on selecting high quality multi-word terms from human written summaries to generate suitable results for web-page summarization.

Content

A Thesis presented to The University of Guelph In partial fulfilment of requirements for the degree of Master of Science in Computer Science. Vgl. Unter: http://www.inf.ufrgs.br%2F~ceramisch%2Fdownload_files%2Fpublications%2F2009%2Fp01.pdf.

Date

10. 1.2013 19:22:47

0.0020823204 = product of:
  0.012493922 = sum of:
    0.012493922 = weight(_text_:in in 1536) [ClassicSimilarity], result of:
      0.012493922 = score(doc=1536,freq=32.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.21040362 = fieldWeight in 1536, product of:
          5.656854 = tf(freq=32.0), with freq of:
            32.0 = termFreq=32.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.02734375 = fieldNorm(doc=1536)
  0.16666667 = coord(1/6)

Abstract

Multiword expressions (MWEs) are lexical items that can be decomposed into single words and display lexical, syntactic, semantic, pragmatic and/or statistical idiosyncrasy (Sag et al., 2002; Kim, 2008; Calzolari et al., 2002). The proper treatment of multiword expressions such as rock 'n' roll and make a decision is essential for many natural language processing (NLP) applications like information extraction and retrieval, terminology extraction and machine translation, and it is important to identify multiword expressions in context. For example, in machine translation we must know that MWEs form one semantic unit, hence their parts should not be translated separately. For this, multiword expressions should be identified first in the text to be translated. The chief aim of this thesis is to develop machine learning-based approaches for the automatic detection of different types of multiword expressions in English and Hungarian natural language texts. In our investigations, we pay attention to the characteristics of different types of multiword expressions such as nominal compounds, multiword named entities and light verb constructions, and we apply novel methods to identify MWEs in raw texts. In the thesis it will be demonstrated that nominal compounds and multiword amed entities may require a similar approach for their automatic detection as they behave in the same way from a linguistic point of view. Furthermore, it will be shown that the automatic detection of light verb constructions can be carried out using two effective machine learning-based approaches.
In this thesis, we focused on the automatic detection of multiword expressions in natural language texts. On the basis of the main contributions, we can argue that: - Supervised machine learning methods can be successfully applied for the automatic detection of different types of multiword expressions in natural language texts. - Machine learning-based multiword expression detection can be successfully carried out for English as well as for Hungarian. - Our supervised machine learning-based model was successfully applied to the automatic detection of nominal compounds from English raw texts. - We developed a Wikipedia-based dictionary labeling method to automatically detect English nominal compounds. - A prior knowledge of nominal compounds can enhance Named Entity Recognition, while previously identified named entities can assist the nominal compound identification process. - The machine learning-based method can also provide acceptable results when it was trained on an automatically generated silver standard corpus. - As named entities form one semantic unit and may consist of more than one word and function as a noun, we can treat them in a similar way to nominal compounds. - Our sequence labelling-based tool can be successfully applied for identifying verbal light verb constructions in two typologically different languages, namely English and Hungarian. - Domain adaptation techniques may help diminish the distance between domains in the automatic detection of light verb constructions. - Our syntax-based method can be successfully applied for the full-coverage identification of light verb constructions. As a first step, a data-driven candidate extraction method can be utilized. After, a machine learning approach that makes use of an extended and rich feature set selects LVCs among extracted candidates. - When a precise syntactic parser is available for the actual domain, the full-coverage identification can be performed better. In other cases, the usage of the sequence labeling method is recommended.

0.0017848461 = product of:
  0.010709076 = sum of:
    0.010709076 = weight(_text_:in in 5976) [ClassicSimilarity], result of:
      0.010709076 = score(doc=5976,freq=32.0), product of:
        0.059380736 = queryWeight, product of:
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.043654136 = queryNorm
        0.18034597 = fieldWeight in 5976, product of:
          5.656854 = tf(freq=32.0), with freq of:
            32.0 = termFreq=32.0
          1.3602545 = idf(docFreq=30841, maxDocs=44218)
          0.0234375 = fieldNorm(doc=5976)
  0.16666667 = coord(1/6)

Abstract

Natural language contradictions are of complex nature. As will be shown in Chapter 5, the realization of contradictions is not limited to the examples such as Socrates is a man and Socrates is not a man (under the condition that Socrates refers to the same object in the real world), which is discussed by Aristotle (Section 3.1.1). Empirical evidence (see Chapter 5 for more details) shows that only a few contradictions occurring in the real life are of that explicit (prototypical) kind. Rather, con-tradictions make use of a variety of natural language devices such as, e.g., paraphrasing, synonyms and antonyms, passive and active voice, diversity of negation expression, and figurative linguistic means such as idioms, irony, and metaphors. Additionally, the most so-phisticated kind of contradictions, the so-called implicit contradictions, can be found only when applying world knowledge and after conducting a sequence of logical operations such as e.g. in: (1.1) The first prize was given to the experienced grandmaster L. Stein who, in total, col-lected ten points (7 wins and 3 draws). Those familiar with the chess rules know that a chess player gets one point for winning and zero points for losing the game. In case of a draw, each player gets a half point. Built on this idea and by conducting some simple mathematical operations, we can infer that in the case of 7 wins and 3 draws (the second part of the sentence), a player can only collect 8.5 points and not 10 points. Hence, we observe that there is a contradiction between the first and the second parts of the sentence.
Implicit contradictions will only partially be the subject of the present study, aiming primarily at identifying the realization mechanism and cues (Chapter 5) as well as finding the parts of contradictions by applying the state of the art algorithms for natural language processing without conducting deep meaning processing. Further in focus are the explicit and implicit contradictions that can be detected by means of explicit linguistic, structural, lexical cues, and by conducting some additional processing operations (e.g., counting the sum in order to detect contradictions arising from numerical divergencies). One should note that an additional complexity in finding contradictions can arise in case parts of the contradictions occur on different levels of realization. Thus, a contradiction can be observed on the word- and phrase-level, such as in a married bachelor (for variations of contradictions on lexical level, see Ganeev 2004), on the sentence level - between parts of a sentence or between two or more sentences, or on the text level - between the portions of a text or between the whole texts such as a contradiction between the Bible and the Quran, for example. Only contradictions arising at the level of single sentences occurring in one or more texts, as well as parts of a sentence, will be considered for the purpose of this study. Though the focus of interest will be on single sentences, it will make use of text particularities such as coreference resolution without establishing the referents in the real world. Finally, another aspect to be considered is that parts of the contradictions are not neces-sarily to appear at the same time. They can be separated by many years and centuries with or without time expression making their recognition by human and detection by machine challenging. According to Aristotle's ontological version of the LNC (Section 3.1.1), how-ever, the same time reference is required in order for two statements to be judged as a contradiction. Taking this into account, we set the borders for the study by limiting the ana-lyzed textual data thematically (only nine world events) and temporally (three days after the reported event had happened) (Section 5.1). No sophisticated time processing will thus be conducted.