Search (16 results, page 1 of 1)

0.025030928 = product of:
  0.050061855 = sum of:
    0.050061855 = sum of:
      0.021780973 = weight(_text_:systems in 1507) [ClassicSimilarity], result of:
        0.021780973 = score(doc=1507,freq=2.0), product of:
          0.16037072 = queryWeight, product of:
            3.0731742 = idf(docFreq=5561, maxDocs=44218)
            0.052184064 = queryNorm
          0.1358164 = fieldWeight in 1507, 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=1507)
      0.028280882 = weight(_text_:22 in 1507) [ClassicSimilarity], result of:
        0.028280882 = score(doc=1507,freq=2.0), product of:
          0.1827397 = queryWeight, product of:
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.052184064 = queryNorm
          0.15476047 = fieldWeight in 1507, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.03125 = fieldNorm(doc=1507)
  0.5 = coord(1/2)

Abstract

Wir werden einmal die Grundlagen des Text-Mining-Systems bei IBM darstellen, dann werden wir das Projekt etwas umfangreicher und deutlicher darstellen, da kennen wir uns aus. Von daher haben wir zwei Teile, einmal Heidelberg, einmal Hamburg. Noch einmal zur Technologie. Text-Mining ist eine von IBM entwickelte Technologie, die in einer besonderen Ausformung und Programmierung für uns zusammengestellt wurde. Das Projekt hieß bei uns lange Zeit DocText Miner und heißt seit einiger Zeit auf Vorschlag von IBM DocCat, das soll eine Abkürzung für Document-Categoriser sein, sie ist ja auch nett und anschaulich. Wir fangen an mit Text-Mining, das bei IBM in Heidelberg entwickelt wurde. Die verstehen darunter das automatische Indexieren als eine Instanz, also einen Teil von Text-Mining. Probleme werden dabei gezeigt, und das Text-Mining ist eben eine Methode zur Strukturierung von und der Suche in großen Dokumentenmengen, die Extraktion von Informationen und, das ist der hohe Anspruch, von impliziten Zusammenhängen. Das letztere sei dahingestellt. IBM macht das quantitativ, empirisch, approximativ und schnell. das muss man wirklich sagen. Das Ziel, und das ist ganz wichtig für unser Projekt gewesen, ist nicht, den Text zu verstehen, sondern das Ergebnis dieser Verfahren ist, was sie auf Neudeutsch a bundle of words, a bag of words nennen, also eine Menge von bedeutungstragenden Begriffen aus einem Text zu extrahieren, aufgrund von Algorithmen, also im Wesentlichen aufgrund von Rechenoperationen. Es gibt eine ganze Menge von linguistischen Vorstudien, ein wenig Linguistik ist auch dabei, aber nicht die Grundlage der ganzen Geschichte. Was sie für uns gemacht haben, ist also die Annotierung von Pressetexten für unsere Pressedatenbank. Für diejenigen, die es noch nicht kennen: Gruner + Jahr führt eine Textdokumentation, die eine Datenbank führt, seit Anfang der 70er Jahre, da sind z.Z. etwa 6,5 Millionen Dokumente darin, davon etwas über 1 Million Volltexte ab 1993. Das Prinzip war lange Zeit, dass wir die Dokumente, die in der Datenbank gespeichert waren und sind, verschlagworten und dieses Prinzip haben wir auch dann, als der Volltext eingeführt wurde, in abgespeckter Form weitergeführt. Zu diesen 6,5 Millionen Dokumenten gehören dann eben auch ungefähr 10 Millionen Faksimileseiten, weil wir die Faksimiles auch noch standardmäßig aufheben.

Date

22. 4.2003 11:45:36

0.019058352 = product of:
  0.038116705 = sum of:
    0.038116705 = product of:
      0.07623341 = sum of:
        0.07623341 = weight(_text_:systems in 3835) [ClassicSimilarity], result of:
          0.07623341 = score(doc=3835,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.47535738 = fieldWeight in 3835, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.109375 = fieldNorm(doc=3835)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Source

Knowledge-based systems. 14(2001) nos.1/2, S.37-53

0.017960707 = product of:
  0.035921413 = sum of:
    0.035921413 = sum of:
      0.021780973 = weight(_text_:systems in 1789) [ClassicSimilarity], result of:
        0.021780973 = score(doc=1789,freq=8.0), product of:
          0.16037072 = queryWeight, product of:
            3.0731742 = idf(docFreq=5561, maxDocs=44218)
            0.052184064 = queryNorm
          0.1358164 = fieldWeight in 1789, product of:
            2.828427 = tf(freq=8.0), with freq of:
              8.0 = termFreq=8.0
            3.0731742 = idf(docFreq=5561, maxDocs=44218)
            0.015625 = fieldNorm(doc=1789)
      0.014140441 = weight(_text_:22 in 1789) [ClassicSimilarity], result of:
        0.014140441 = score(doc=1789,freq=2.0), product of:
          0.1827397 = queryWeight, product of:
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.052184064 = queryNorm
          0.07738023 = fieldWeight in 1789, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.015625 = fieldNorm(doc=1789)
  0.5 = coord(1/2)

Date

23. 3.2008 19:10:22

Footnote

In 13 chapters, Part Two provides an introduction to KDD, an overview of data mining techniques, and examples of the usefulness of data model visualizations. The importance of visualization throughout the KDD process is stressed in many of the chapters. In particular, the need for measures of visualization effectiveness, benchmarking for identifying best practices, and the use of standardized sample data sets is convincingly presented. Many of the important data mining approaches are discussed in this complementary context. Cluster and outlier detection, classification techniques, and rule discovery algorithms are presented as the basic techniques common to the KDD process. The potential effectiveness of using visualization in the data modeling process are illustrated in chapters focused an using visualization for helping users understand the KDD process, ask questions and form hypotheses about their data, and evaluate the accuracy and veracity of their results. The 11 chapters of Part Three provide an overview of the KDD process and successful approaches to integrating KDD, data mining, and visualization in complementary domains. Rhodes (Chapter 21) begins this section with an excellent overview of the relation between the KDD process and data mining techniques. He states that the "primary goals of data mining are to describe the existing data and to predict the behavior or characteristics of future data of the same type" (p. 281). These goals are met by data mining tasks such as classification, regression, clustering, summarization, dependency modeling, and change or deviation detection. Subsequent chapters demonstrate how visualization can aid users in the interactive process of knowledge discovery by graphically representing the results from these iterative tasks. Finally, examples of the usefulness of integrating visualization and data mining tools in the domain of business, imagery and text mining, and massive data sets are provided. This text concludes with a thorough and useful 17-page index and lengthy yet integrating 17-page summary of the academic and industrial backgrounds of the contributing authors. A 16-page set of color inserts provide a better representation of the visualizations discussed, and a URL provided suggests that readers may view all the book's figures in color on-line, although as of this submission date it only provides access to a summary of the book and its contents. The overall contribution of this work is its focus an bridging two distinct areas of research, making it a valuable addition to the Morgan Kaufmann Series in Database Management Systems. The editors of this text have met their main goal of providing the first textbook integrating knowledge discovery, data mining, and visualization. Although it contributes greatly to our under- standing of the development and current state of the field, a major weakness of this text is that there is no concluding chapter to discuss the contributions of the sum of these contributed papers or give direction to possible future areas of research. "Integration of expertise between two different disciplines is a difficult process of communication and reeducation. Integrating data mining and visualization is particularly complex because each of these fields in itself must draw an a wide range of research experience" (p. 300). Although this work contributes to the crossdisciplinary communication needed to advance visualization in KDD, a more formal call for an interdisciplinary research agenda in a concluding chapter would have provided a more satisfying conclusion to a very good introductory text.

LCSH

Knowledge acquisition (Expert systems)

Series

Morgan Kaufmann series in data management systems

Subject

Knowledge acquisition (Expert systems)

0.011551105 = product of:
  0.02310221 = sum of:
    0.02310221 = product of:
      0.04620442 = sum of:
        0.04620442 = weight(_text_:systems in 4863) [ClassicSimilarity], result of:
          0.04620442 = score(doc=4863,freq=4.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.28811008 = fieldWeight in 4863, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.046875 = fieldNorm(doc=4863)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

In data mining, we emphasize the need for learning from huge, incomplete, and imperfect data sets. To handle noise in the problem domain, existing learning systems avoid overfitting the imperfect training examples by excluding insignificant patterns. The problem is that these systems use a limiting attribute-value language for representing the training examples and the induced knowledge. Moreover, some important patterns are ignored because they are statistically insignificant. In this article, we present a framework that combines genetic programming and inductive logic programming to induce knowledge represented in various knowledge representation formalisms from noisy databases (LOGENPRO). Moreover, the system is applied to one real-life medical database. The knowledge discovered provides insights to and allows better understanding of the medical domains

0.011551105 = product of:
  0.02310221 = sum of:
    0.02310221 = product of:
      0.04620442 = sum of:
        0.04620442 = weight(_text_:systems in 1497) [ClassicSimilarity], result of:
          0.04620442 = score(doc=1497,freq=4.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.28811008 = fieldWeight in 1497, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.046875 = fieldNorm(doc=1497)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Text mining is about making serendipity more likely. Serendipity, the chance discovery of interesting ideas, has been responsible for many discoveries in science. Text mining systems strive to explore large text collections, separate the potentially meaningfull connections from a vast and mostly noisy background of random associations. In this paper we provide a summary of our text mining approach and also illustrate briefly some of the experiments we have conducted with this approach. In particular we use a profile-based text mining method. We have used these profiles to explore the global distribution of disease research, replicate discoveries made by others and propose new hypotheses. Text mining holds much potential that has yet to be tapped.

Source

Knowledge organization, information systems and other essays: Professor A. Neelameghan Festschrift. Ed. by K.S. Raghavan and K.N. Prasad

0.011551105 = product of:
  0.02310221 = sum of:
    0.02310221 = product of:
      0.04620442 = sum of:
        0.04620442 = weight(_text_:systems in 2806) [ClassicSimilarity], result of:
          0.04620442 = score(doc=2806,freq=4.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.28811008 = fieldWeight in 2806, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.046875 = fieldNorm(doc=2806)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

External information plays an important role in group decision-making processes, yet research about external information support for Group Support Systems (GSS) has been lacking. In this study, we propose an approach to build a concept space to provide external concept support for GSS users. Built on a Web mining algorithm, the approach can mine a concept space from the Web and retrieve related concepts from the concept space based on users' comments in a real-time manner. We conduct two experiments to evaluate the quality of the proposed approach and the effectiveness of the external concept support provided by this approach. The experiment results indicate that the concept space mined from the Web contained qualified concepts to stimulate divergent thinking. The results also demonstrate that external concept support in GSS greatly enhanced group productivity for idea generation tasks.

0.010605331 = product of:
  0.021210661 = sum of:
    0.021210661 = product of:
      0.042421322 = sum of:
        0.042421322 = weight(_text_:22 in 1383) [ClassicSimilarity], result of:
          0.042421322 = score(doc=1383,freq=2.0), product of:
            0.1827397 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.052184064 = queryNorm
            0.23214069 = fieldWeight in 1383, 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=1383)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Date

22. 3.2008 14:46:06

0.009625921 = product of:
  0.019251842 = sum of:
    0.019251842 = product of:
      0.038503684 = sum of:
        0.038503684 = weight(_text_:systems in 3237) [ClassicSimilarity], result of:
          0.038503684 = score(doc=3237,freq=4.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.24009174 = fieldWeight in 3237, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0390625 = fieldNorm(doc=3237)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

In this article we present a semi-supervised active learning algorithm for pattern discovery in information extraction from textual data. The patterns are reduced regular expressions composed of various characteristics of features useful in information extraction. Our major contribution is a semi-supervised learning algorithm that extracts information from a set of examples labeled as relevant or irrelevant to a given attribute. The approach is semi-supervised because it does not require precise labeling of the exact location of features in the training data. This significantly reduces the effort needed to develop a training set. An active learning algorithm is used to assist the semi-supervised learning algorithm to further reduce the training set development effort. The active learning algorithm is seeded with a Single positive example of a given attribute. The context of the seed is used to automatically identify candidates for additional positive examples of the given attribute. Candidate examples are manually pruned during the active learning phase, and our semi-supervised learning algorithm automatically discovers reduced regular expressions for each attribute. We have successfully applied this learning technique in the extraction of textual features from police incident reports, university crime reports, and patents. The performance of our algorithm compares favorably with competitive extraction systems being used in criminal justice information systems.

0.0070702205 = product of:
  0.014140441 = sum of:
    0.014140441 = product of:
      0.028280882 = sum of:
        0.028280882 = weight(_text_:22 in 2403) [ClassicSimilarity], result of:
          0.028280882 = score(doc=2403,freq=2.0), product of:
            0.1827397 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.052184064 = queryNorm
            0.15476047 = fieldWeight in 2403, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.03125 = fieldNorm(doc=2403)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Date

3. 5.1997 8:44:22

0.0068065543 = product of:
  0.013613109 = sum of:
    0.013613109 = product of:
      0.027226217 = sum of:
        0.027226217 = weight(_text_:systems in 5083) [ClassicSimilarity], result of:
          0.027226217 = score(doc=5083,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1697705 = fieldWeight in 5083, 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=5083)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

In this paper, we discuss Conceptual Knowledge Discovery in Databases (CKDD) in its connection with Data Analysis. Our approach is based on Formal Concept Analysis, a mathematical theory which has been developed and proven useful during the last 20 years. Formal Concept Analysis has led to a theory of conceptual information systems which has been applied by using the management system TOSCANA in a wide range of domains. In this paper, we use such an application in database marketing to demonstrate how methods and procedures of CKDD can be applied in Data Analysis. In particular, we show the interplay and integration of data mining and data analysis techniques based on Formal Concept Analysis. The main concern of this paper is to explain how the transition from data to knowledge can be supported by a TOSCANA system. To clarify the transition steps we discuss their correspondence to the five levels of knowledge representation established by R. Brachman and to the steps of empirically grounded theory building proposed by A. Strauss and J. Corbin

0.0068065543 = product of:
  0.013613109 = sum of:
    0.013613109 = product of:
      0.027226217 = sum of:
        0.027226217 = weight(_text_:systems in 3603) [ClassicSimilarity], result of:
          0.027226217 = score(doc=3603,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1697705 = fieldWeight in 3603, 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=3603)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Source

Gaining insight from research information (CRIS2002): Proceedings of the 6th International Conference an Current Research Information Systems, University of Kassel, August 29 - 31, 2002. Eds: W. Adamczak u. A. Nase

0.0068065543 = product of:
  0.013613109 = sum of:
    0.013613109 = product of:
      0.027226217 = sum of:
        0.027226217 = weight(_text_:systems in 597) [ClassicSimilarity], result of:
          0.027226217 = score(doc=597,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1697705 = fieldWeight in 597, 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=597)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

With the overwhelming volume of information, the task of finding relevant information on a given topic on the Web is becoming increasingly difficult. Web search engines hence become one of the most popular solutions available on the Web. However, it has never been easy for novice users to organize and represent their information needs using simple queries. Users have to keep modifying their input queries until they get expected results. Therefore, it is often desirable for search engines to give suggestions on related queries to users. Besides, by identifying those related queries, search engines can potentially perform optimizations on their systems, such as query expansion and file indexing. In this work we propose a method that suggests a list of related queries given an initial input query. The related queries are based in the query log of previously submitted queries by human users, which can be identified using an enhanced model of association rules. Users can utilize the suggested related queries to tune or redirect the search process. Our method not only discovers the related queries, but also ranks them according to the degree of their relatedness. Unlike many other rival techniques, it also performs reasonably well on less frequent input queries.

0.0068065543 = product of:
  0.013613109 = sum of:
    0.013613109 = product of:
      0.027226217 = sum of:
        0.027226217 = weight(_text_:systems in 604) [ClassicSimilarity], result of:
          0.027226217 = score(doc=604,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1697705 = fieldWeight in 604, 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=604)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Modern information retrieval systems use keywords within documents as indexing terms for search of relevant documents. As Chinese is an ideographic character-based language, the words in the texts are not delimited by white spaces. Indexing of Chinese documents is impossible without a proper segmentation algorithm. Many Chinese segmentation algorithms have been proposed in the past. Traditional segmentation algorithms cannot operate without a large dictionary or a large corpus of training data. Nowadays, the Web has become the largest corpus that is ideal for Chinese segmentation. Although most search engines have problems in segmenting texts into proper words, they maintain huge databases of documents and frequencies of character sequences in the documents. Their databases are important potential resources for segmentation. In this paper, we propose a segmentation algorithm by mining Web data with the help of search engines. On the other hand, the Romanized pinyin of Chinese language indicates boundaries of words in the text. Our algorithm is the first to utilize the Romanized pinyin to segmentation. It is the first unified segmentation algorithm for the Chinese language from different geographical areas, and it is also domain independent because of the nature of the Web. Experiments have been conducted on the datasets of a recent Chinese segmentation competition. The results show that our algorithm outperforms the traditional algorithms in terms of precision and recall. Moreover, our algorithm can effectively deal with the problems of segmentation ambiguity, new word (unknown word) detection, and stop words.

0.0068065543 = product of:
  0.013613109 = sum of:
    0.013613109 = product of:
      0.027226217 = sum of:
        0.027226217 = weight(_text_:systems in 605) [ClassicSimilarity], result of:
          0.027226217 = score(doc=605,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1697705 = fieldWeight in 605, 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=605)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Documents discussing public affairs, common themes, interesting products, and so on, are reported and distributed on the Web. Positive and negative opinions embedded in documents are useful references and feedbacks for governments to improve their services, for companies to market their products, and for customers to purchase their objects. Web opinion mining aims to extract, summarize, and track various aspects of subjective information on the Web. Mining subjective information enables traditional information retrieval (IR) systems to retrieve more data from human viewpoints and provide information with finer granularity. Opinion extraction identifies opinion holders, extracts the relevant opinion sentences, and decides their polarities. Opinion summarization recognizes the major events embedded in documents and summarizes the supportive and the nonsupportive evidence. Opinion tracking captures subjective information from various genres and monitors the developments of opinions from spatial and temporal dimensions. To demonstrate and evaluate the proposed opinion mining algorithms, news and bloggers' articles are adopted. Documents in the evaluation corpora are tagged in different granularities from words, sentences to documents. In the experiments, positive and negative sentiment words and their weights are mined on the basis of Chinese word structures. The f-measure is 73.18% and 63.75% for verbs and nouns, respectively. Utilizing the sentiment words mined together with topical words, we achieve f-measure 62.16% at the sentence level and 74.37% at the document level.

0.0053026653 = product of:
  0.010605331 = sum of:
    0.010605331 = product of:
      0.021210661 = sum of:
        0.021210661 = weight(_text_:22 in 1178) [ClassicSimilarity], result of:
          0.021210661 = score(doc=1178,freq=2.0), product of:
            0.1827397 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.052184064 = queryNorm
            0.116070345 = fieldWeight in 1178, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0234375 = fieldNorm(doc=1178)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Content

"Ob man als Terrorist einen Anschlag gegen die Vereinigten Staaten plant, als Kassierer Scheine aus der Kasse unterschlägt oder für bestimmte Produkte besonders gerne Geld ausgibt - einen Unterschied macht Data-Mining-Software da nicht. Solche Programme analysieren riesige Daten- mengen und fällen statistische Urteile. Mit diesen Methoden wollen nun die For- scher des "Information Awaren in den Vereinigten Staaten Spuren von Terroristen in den Datenbanken von Behörden und privaten Unternehmen wie Kreditkartenfirmen finden. 200 Millionen Dollar umfasst der Jahresetat für die verschiedenen Forschungsprojekte. Dass solche Software in der Praxis funktioniert, zeigen die steigenden Umsätze der Anbieter so genannter Customer-Relationship-Management-Software. Im vergangenen Jahr ist das Potenzial für analytische CRM-Anwendungen laut dem Marktforschungsinstitut IDC weltweit um 22 Prozent gewachsen, bis zum Jahr 2006 soll es in Deutschland mit einem jährlichen Plus von 14,1 Prozent so weitergehen. Und das trotz schwacher Konjunktur - oder gerade deswegen. Denn ähnlich wie Data-Mining der USRegierung helfen soll, Terroristen zu finden, entscheiden CRM-Programme heute, welche Kunden für eine Firma profitabel sind. Und welche es künftig sein werden, wie Manuela Schnaubelt, Sprecherin des CRM-Anbieters SAP, beschreibt: "Die Kundenbewertung ist ein zentraler Bestandteil des analytischen CRM. Sie ermöglicht es Unternehmen, sich auf die für sie wichtigen und richtigen Kunden zu fokussieren. Darüber hinaus können Firmen mit speziellen Scoring- Verfahren ermitteln, welche Kunden langfristig in welchem Maße zum Unternehmenserfolg beitragen." Die Folgen der Bewertungen sind für die Betroffenen nicht immer positiv: Attraktive Kunden profitieren von individuellen Sonderangeboten und besonderer Zuwendung. Andere hängen vielleicht so lauge in der Warteschleife des Telefonservice, bis die profitableren Kunden abgearbeitet sind. So könnte eine praktische Umsetzung dessen aussehen, was SAP-Spreche-rin Schnaubelt abstrakt beschreibt: "In vielen Unternehmen wird Kundenbewertung mit der klassischen ABC-Analyse durchgeführt, bei der Kunden anhand von Daten wie dem Umsatz kategorisiert werden. A-Kunden als besonders wichtige Kunden werden anders betreut als C-Kunden." Noch näher am geplanten Einsatz von Data-Mining zur Terroristenjagd ist eine Anwendung, die heute viele Firmen erfolgreich nutzen: Sie spüren betrügende Mitarbeiter auf. Werner Sülzer vom großen CRM-Anbieter NCR Teradata beschreibt die Möglichkeiten so: "Heute hinterlässt praktisch jeder Täter - ob Mitarbeiter, Kunde oder Lieferant - Datenspuren bei seinen wirtschaftskriminellen Handlungen. Es muss vorrangig darum gehen, einzelne Spuren zu Handlungsmustern und Täterprofilen zu verdichten. Das gelingt mittels zentraler Datenlager und hoch entwickelter Such- und Analyseinstrumente." Von konkreten Erfolgen sprich: Entlas-sungen krimineller Mitarbeiter-nach Einsatz solcher Programme erzählen Unternehmen nicht gerne. Matthias Wilke von der "Beratungsstelle für Technologiefolgen und Qualifizierung" (BTQ) der Gewerkschaft Verdi weiß von einem Fall 'aus der Schweiz. Dort setzt die Handelskette "Pick Pay" das Programm "Lord Lose Prevention" ein. Zwei Monate nach Einfüh-rung seien Unterschlagungen im Wert von etwa 200 000 Franken ermittelt worden. Das kostete mehr als 50 verdächtige Kassiererinnen und Kassierer den Job.

0.0053026653 = product of:
  0.010605331 = sum of:
    0.010605331 = product of:
      0.021210661 = sum of:
        0.021210661 = weight(_text_:22 in 1833) [ClassicSimilarity], result of:
          0.021210661 = score(doc=1833,freq=2.0), product of:
            0.1827397 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.052184064 = queryNorm
            0.116070345 = fieldWeight in 1833, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0234375 = fieldNorm(doc=1833)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Date

11. 5.2008 19:49:22