Diese Datenbank enthält über 40.000 Dokumente zu Themen aus den Bereichen Formalerschließung – Inhaltserschließung – Information Retrieval.
© 2015 W. Gödert, TH Köln, Institut für Informationswissenschaft / Powered by litecat, BIS Oldenburg (Stand: 04. Juni 2021)
1Saarikoski, J. ; Laurikkala, J. ; Järvelin, K. ; Juhola, M.: ¬A study of the use of self-organising maps in information retrieval.
In: Journal of documentation. 65(2009) no.2, S.304-322.
Abstract: Purpose - The aim of this paper is to explore the possibility of retrieving information with Kohonen self-organising maps, which are known to be effective to group objects according to their similarity or dissimilarity. Design/methodology/approach - After conventional preprocessing, such as transforming into vector space, documents from a German document collection were trained for a neural network of Kohonen self-organising map type. Such an unsupervised network forms a document map from which relevant objects can be found according to queries. Findings - Self-organising maps ordered documents to groups from which it was possible to find relevant targets. Research limitations/implications - The number of documents used was moderate due to the limited number of documents associated to test topics. The training of self-organising maps entails rather long running times, which is their practical limitation. In future, the aim will be to build larger networks by compressing document matrices, and to develop document searching in them. Practical implications - With self-organising maps the distribution of documents can be visualised and relevant documents found in document collections of limited size. Originality/value - The paper reports on an approach that can be especially used to group documents and also for information search. So far self-organising maps have rarely been studied for information retrieval. Instead, they have been applied to document grouping tasks.
2Subramanyam Rallabandi, V.P. ; Sett, S.K.: Knowledge-based image retrieval system.
In: Knowledge-based systems. 21(2008) no.2, S.89-100.
Abstract: Most of the retrieval systems concentrate much on low-level features such as color, texture, shape and position. The present system is mainly developed based on the visual descriptors of the image such as color, texture and shape descriptors, etc. along with the high-level semantic analysis of the image content through different processing modules in the proposed architecture. Similarity measures are proposed and the performance evaluation has been done. As an image browser, apart from retrieving images by image example, it also supports query by natural language. The present system works well both online and offline.We have used unsupervised Kohonen's self-organizing maps (SOM) technique to train the images and our own indexing scheme with reference system based on R-tree SOM. We proposed an approach fuzzy color histogram for color retrieval, and Lie descriptors for the retrieval of shapes. We have also tested our appraoch for MPEG-7 images.
Behandelte Form: Bilder
3Ding, C. ; Patra, J.C.: User modeling for personalized Web search with Self-Organizing Map.
In: Journal of the American Society for Information Science and Technology. 58(2007) no.4, S.494-504.
Abstract: The widely used Web search engines index and recommend individual Web pages in response to a few keywords queries to assist users in locating relevant documents. However, the Web search engines give different users the same answer set, although the users may have different preferences. A personalized Web search would carry out the search for each user according to his or her preferences. To conduct the personalized Web search, the authors provide a novel approach to model the user profile with a self-organizing map (SOM). Their results indicate that SOM is capable of helping the user to find the related category for each query used in the Web search to make a personalized Web search effective.
4Goren-Bar, D. ; Kuflik, T.: Supporting user-subjective categorization with self-organizing maps and learning vector quantization.
In: Journal of the American Society for Information Science and Technology. 56(2005) no.4, S.345-355.
Abstract: Today, most document categorization in organizations is done manually. We save at work hundreds of files and e-mail messages in folders every day. While automatic document categorization has been widely studied, much challenging research still remains to support usersubjective categorization. This study evaluates and compares the application of self-organizing maps (SOMs) and learning vector quantization (LVO) with automatic document classification, using a set of documents from an organization, in a specific domain, manually classified by a domain expert. After running the SOM and LVO we requested the user to reclassify documents that were misclassified by the system. Results show that despite the subjective nature of human categorization, automatic document categorization methods correlate weIl with subjective, personal categorization, and the LVO method outperforms the SOM. The reclassification process revealed an interesting pattern: About 40% of the documents were classified according to their original categorization, about 35% according to the system's categorization (the users changed the original categorization), and the remainder received a different (new) categorization. Based an these results we conclude that automatic support for subjective categorization is feasible; however, an exact match is probably impossible due to the users' changing categorization behavior.
5Faba-Perez, C. ; Guerrero-Bote, V.P. ; Moya-Anegon, F. de: Self-organizing maps of Web spaces based an formal characteristics.
In: Information processing and management. 41(2005) no.2, S.331-346.
Abstract: The unceasing growth of electronic information available on the Web has made it indispensable to develop tools to analyse and evaluate its quality. Given the subjectivity underlying most of the qualitative analytical indicators at the present time, we here propose the use of characteristics or indicators of a formal character. We apply Kohonen's neural networks and study their topological organization in order to analyse how Web spaces behave with respect to these networks' formal characteristics. The interpretation of the results brings out the underlying structures and relationships in a closed Web environment.
6Chen, H. ; Lally, A.M. ; Zhu, B. ; Chau, M.: HelpfulMed : Intelligent searching for medical information over the Internet.
In: Journal of the American Society for Information Science and technology. 54(2003) no.7, S.683-694.
Abstract: The Medical professionals and researchers need information from reputable sources to accomplish their work. Unfortunately, the Web has a large number of documents that are irrelevant to their work, even those documents that purport to be "medically-related." This paper describes an architecture designed to integrate advanced searching and indexing algorithms, an automatic thesaurus, or "concept space," and Kohonen-based Self-Organizing Map (SOM) technologies to provide searchers with finegrained results. Initial results indicate that these systems provide complementary retrieval functionalities. HelpfulMed not only allows users to search Web pages and other online databases, but also allows them to build searches through the use of an automatic thesaurus and browse a graphical display of medical-related topics. Evaluation results for each of the different components are included. Our spidering algorithm outperformed both breadth-first search and PageRank spiders an a test collection of 100,000 Web pages. The automatically generated thesaurus performed as well as both MeSH and UMLS-systems which require human mediation for currency. Lastly, a variant of the Kohonen SOM was comparable to MeSH terms in perceived cluster precision and significantly better at perceived cluster recall.
Anmerkung: Teil eines Themenheftes: "Web retrieval and mining: A machine learning perspective"
Themenfeld: Semantisches Umfeld in Indexierung u. Retrieval ; Retrievalalgorithmen
Objekt: HelpfulMed ; SOM ; MeSH ; UMLS
7red: Alles Wissen gleich einer großen Stadt.
In: Der Standard. Nr.xx vom 9.11.2002, S.x.
Inhalt: "Das rasant wachsende Wissen muss gut verwaltet werden, um es zu nutzen. Dies erfordert intelligente Wissensmanagementsysteme, wie sie Andreas Rauber von der Technischen Uni Wien über digitale Bibliotheken konzipiert hat. Seine "Wissenslandkarte" erlaubt es, große Datenmengen übersichtlich darzustellen, Wissen rasch auffindbar und damit optimal einsetzbar zu machen. Dafür erhielt er nun den Cor Baayen Award 2002 für aussichtsreiche Nachwuchsforscher im Bereich der Informationstechnologie vom European Research Consortium for Informatics and Mathematics. Rauber entwickelte eine Bibliothek, die auf einer sich selbst organisierenden Landkarte basiert: Einer geographischen Landkarte gleich, ist themenverwandtes Wissen in Form eines Clusters abgebildet, quasi als städtischer Ballungsraum. Damit verbundene Inhalte sind räumlich gesehen in kurzer Distanz dazu abgebildet, vergleichbar den Randgebieten des Ballungsraumes. So ist auf einen Blick ersichtlich, wo bestimmte Themenkomplexe und damit verbundene Inhalte in der Bibliothek abgelegt sind. Die Wissenslandkarte bedient sich der Forschungen zu neuronalen Netzen. Durch ein Verfahren erlernt die "Self-Organizing-Map" (SOM) die Inhalte der einzelnen Dokumente und schafft es, mit zunehmender Datenmenge selbst eine Struktur des vorhandenen Wissens zu erstellen. Dieses Verfahren ist sprachunabhängig und daher weltweit einsetzbar."
Themenfeld: Semantisches Umfeld in Indexierung u. Retrieval
8Wu, Q. ; Iyengar, S.S. ; Zhu, M.: Web based image retrieval using self-organizing feature map.
In: Journal of the American Society for Information Science and technology. 52(2001) no.10, S.868-875.
Abstract: The explosive growth of digital image collections on the Web sites is calling for an efficient and intelligent method of browsing, searching, and retrieving images. In this article, an artificial neural network (ANN)-based approach is proposed to explore a promising solution to the Web image retrieval (IR). Compared with other image retrieval methods, this new approach has the following characteristics. First of all, the Content-Based features have been combined with Text-Based features to improve retrieval performance. Instead of solely relying on low-level visual features and high-level concepts, we also take the textual features into consideration, which are automatically extracted from image names, alternative names, page titles, surrounding texts, URLs, etc. Secondly, the Kohonen neural network model is introduced and led into the image retrieval process. Due to its self-organizing property, the cognitive knowledge is learned, accumulated, and solidified during the unsupervised training process. The architecture is presented to illustrate the main conceptual components and mechanism of the proposed image retrieval system. To demonstrate the superiority of the new IR system over other IR systems, the retrieval result of a test example is also given in the article.
Behandelte Form: Bilder
Objekt: Kohonen-Netz ; SOM
9Guerrero, V.P. ; Moya Anegón, F. de: Reduction of the dimension of a document space using the fuzzified output of a Kohonen network.
In: Journal of the American Society for Information Science and technology. 52(2001) no.14, S.1234-1241.
Abstract: The vectors used in IR, whether to represent the documents or the terms, are high dimensional, and their dimensions increase as one approaches real problems. The algorithms used to manipulate them, however, consume enormously increasing amounts of computational capacity as the said dimension grows. We used the Kohonen algorithm and a fuzzification module to perform a fuzzy clustering of the terms. The degrees of membership obtained were used to represent the terms and, by extension, the documents, yielding a smaller number of components but still endowed with meaning. To test the results, we use a topological classification of sets of transformed and untransformed vectors to check that the same structure underlies both.
Objekt: Kohonen-Netz ; SOM
10Chen, H. ; Houston, A.L. ; Sewell, R.R. ; Schatz, B.R.: Internet browsing and searching : user evaluations of category map and concept space techniques.
In: Journal of the American Society for Information Science. 49(1998) no.7, S.582-603.
Abstract: The Internet provides an exceptional testbed for developing algorithms that can improve bowsing and searching large information spaces. Browsing and searching tasks are susceptible to problems of information overload and vocabulary differences. Much of the current research is aimed at the development and refinement of algorithms to improve browsing and searching by addressing these problems. Our research was focused on discovering whether two of the algorithms our research group has developed, a Kohonen algorithm category map for browsing, and an automatically generated concept space algorithm for searching, can help improve browsing and / or searching the Internet. Our results indicate that a Kohonen self-organizing map (SOM)-based algorithm can successfully categorize a large and eclectic Internet information space (the Entertainment subcategory of Yahoo!) into manageable sub-spaces that users can successfully navigate to locate a homepage of interest to them. The SOM algorithm worked best with browsing tasks that were very broad, and in which subjects skipped around between categories. Subjects especially liked the visual and graphical aspects of the map. Subjects who tried to do a directed search, and those that wanted to use the more familiar mental models (alphabetic or hierarchical organization) for browsing, found that the work did not work well. The results from the concept space experiment were especially encouraging. There were no significant differences among the precision measures for the set of documents identified by subject-suggested terms, thesaurus-suggested terms, and the combination of subject- and thesaurus-suggested terms. The recall measures indicated that the combination of subject- and thesaurs-suggested terms exhibited significantly better recall than subject-suggested terms alone. Furthermore, analysis of the homepages indicated that there was limited overlap between the homepages retrieved by the subject-suggested and thesaurus-suggested terms. Since the retrieval homepages for the most part were different, this suggests that a user can enhance a keyword-based search by using an automatically generated concept space. Subejcts especially liked the level of control that they could exert over the search, and the fact that the terms suggested by the thesaurus were 'real' (i.e., orininating in the homepages) and therefore guaranteed to have retrieval success
Objekt: Kohonen-Netz ; SOM
11Orwig, R.E. ; Chen, H. ; Nunamaker, J.F.: ¬A graphical, self-organizing approach to classifying electronic meeting output.
In: Journal of the American Society for Information Science. 48(1997) no.2, S.157-170.
Abstract: Describes research in the application of a Kohonen Self-Organizing Map (SOM) to the problem of classification of electronic brainstorming output and an evaluation of the results. Describes an electronic meeting system and describes the classification problem that exists in the group problem solving process. Surveys the literature concerning classification. Describes the application of the Kohonen SOM to the meeting output classification problem. Describes an experiment that evaluated the classification performed by the Kohonen SOM by comparing it with those of a human expert and a Hopfield neural network. Discusses conclusions and directions for future research
Themenfeld: Automatisches Klassifizieren
Objekt: Hopfield-Netze ; Kohonen-Netz ; SOM
12Zavrel, J.: Neural navigation interfaces for information retrieval : are they more than an appealing idea?.
In: Artificial intelligence review. 10(1996) nos.5/6, S.477-504.
Abstract: Gives an overview of research in the area of using neural networks to construct navigation interfaces for information retrieval systems. Identifies problems in the application of Kohonen networks for information retrieval, and proposes to use the growing cell structures network, a variant of the Kohonen network which shows a more flexible adaptation to the domain structure. This network was tested on 2 standard test collections, using a combined recall and precision measure, and compared to traditional information retrieval methods. It performs at a competitive level of effectiveness, and is suitable for visualisation purposes. However, the incremental training procedures for networks result in a reliability problem, and the approach is computationally intensive. Also, the utility of the rsulting maps for navigation will need further improvement
Anmerkung: Contribution to a special issue on the application of artificial intelligence to information retrieval
Objekt: Kohonen-Netz ; SOM
14Feiten, B. ; Gunzel, S.: Automatic indexing of a sound database using self-organizing neural nets.
In: Computer music journal. 18(1994) no.3, S.53-65.
Abstract: Modern techniques of electronic sound synthesis and computer based sound archiving place a large number of various sounds at the musician's disposal. However, as the number of available sounds increases, so too does the time and effort required to select a sound. Describes results of research which uses the organization of sounds based on a neural network called Kohonen feature map to produce a retrieval index for sound archives automatically. The adaptation of the metrics for the sound space to the auditory sensation leads to a retrieval index that provides virtual perception. Similarities of sounds can be evaluated mathematically
Behandelte Form: Musik-Tonträger
Objekt: Kohonen-Netz ; SOM