NN approaches to natural language: Context and trends

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review


    This is a short introduction to topics and problems in natural language processing that are of interest for connectionist research. We recall the natural language processing (NLP) problem and conclude that until now no promising approaches from a neural network (NN) point of view are available. It does not mean that there is no progress in this field. Rather than attacking the complete problem it has become customary to search for subtasks for which a NN approach is useful. In addition, the sometimes rather ad hoc and exotic approaches to language analysis have been replaced by fundamental research into sequence recognition, finite state automata (FSA) simulation and general research on grammatical inference with NNs. We do not confine ourselves to NN approaches, but survey a range of approaches to NLP and provide the framework in which connectionist NLP can be embedded and evaluated.
    Original languageEnglish
    Title of host publicationICANN ’93
    Subtitle of host publicationProceedings of the International Conference on Artificial Neural Networks Amsterdam, The Netherlands 13–16 September 1993
    EditorsStan Gielen, Bert Kappen
    Place of PublicationLondon
    Number of pages8
    ISBN (Electronic)978-1-4471-2063-6
    ISBN (Print)978-3-540-19839-0
    Publication statusPublished - 13 Sept 1993
    EventInternational Conference on Artificial Neural Networks, ICANN 1993 - Amsterdam, Netherlands
    Duration: 13 Sept 199316 Sept 1993


    ConferenceInternational Conference on Artificial Neural Networks, ICANN 1993
    Abbreviated titleICANN


    • HMI-CI: Computational Intelligence
    • Natural language processing
    • Neural network approach
    • Parse tree
    • Word sense disambiguation
    • Finite state automaton


    Dive into the research topics of 'NN approaches to natural language: Context and trends'. Together they form a unique fingerprint.

    Cite this