A note on morphological analysis methods based on statistical decision theory

Yasunari Maeda, Naoya Ikeda, Hideki Yoshida, Yoshitaka Fujiwara, Toshiyasu Matsushima

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Morphological analysis is one of important topics in the field of NLP(Natural Language Processing). In many previous research a HMM(Hidden Markov Model) with unknown parameters has been used as a language model. In this research we also use the HMM as the language model. And we assume that sate transitions in the HMM are dominated by a second order Markov chain. At first we propose two types of morphological analysis methods which minimize the error rate with reference to a Bayes criterion. But the computational complexity of the proposed Bayes optimal morphological analysis methods are exponential order. So we also propose approximate methods.

    Original languageEnglish
    Title of host publicationProceedings of the SICE Annual Conference
    Pages1563-1568
    Number of pages6
    DOIs
    Publication statusPublished - 2007
    EventSICE(Society of Instrument and Control Engineers)Annual Conference, SICE 2007 - Takamatsu
    Duration: 2007 Sep 172007 Sep 20

    Other

    OtherSICE(Society of Instrument and Control Engineers)Annual Conference, SICE 2007
    CityTakamatsu
    Period07/9/1707/9/20

    Fingerprint

    Decision theory
    Hidden Markov models
    Markov processes
    Computational complexity
    Processing

    Keywords

    • Hidden markov model
    • Morphological analysis
    • Statistical decision theory

    ASJC Scopus subject areas

    • Engineering(all)

    Cite this

    Maeda, Y., Ikeda, N., Yoshida, H., Fujiwara, Y., & Matsushima, T. (2007). A note on morphological analysis methods based on statistical decision theory. In Proceedings of the SICE Annual Conference (pp. 1563-1568). [4421232] https://doi.org/10.1109/SICE.2007.4421232

    A note on morphological analysis methods based on statistical decision theory. / Maeda, Yasunari; Ikeda, Naoya; Yoshida, Hideki; Fujiwara, Yoshitaka; Matsushima, Toshiyasu.

    Proceedings of the SICE Annual Conference. 2007. p. 1563-1568 4421232.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Maeda, Y, Ikeda, N, Yoshida, H, Fujiwara, Y & Matsushima, T 2007, A note on morphological analysis methods based on statistical decision theory. in Proceedings of the SICE Annual Conference., 4421232, pp. 1563-1568, SICE(Society of Instrument and Control Engineers)Annual Conference, SICE 2007, Takamatsu, 07/9/17. https://doi.org/10.1109/SICE.2007.4421232
    Maeda Y, Ikeda N, Yoshida H, Fujiwara Y, Matsushima T. A note on morphological analysis methods based on statistical decision theory. In Proceedings of the SICE Annual Conference. 2007. p. 1563-1568. 4421232 https://doi.org/10.1109/SICE.2007.4421232
    Maeda, Yasunari ; Ikeda, Naoya ; Yoshida, Hideki ; Fujiwara, Yoshitaka ; Matsushima, Toshiyasu. / A note on morphological analysis methods based on statistical decision theory. Proceedings of the SICE Annual Conference. 2007. pp. 1563-1568
    @inproceedings{2f35e5295d1a4126a8ed1ef053c593f0,
    title = "A note on morphological analysis methods based on statistical decision theory",
    abstract = "Morphological analysis is one of important topics in the field of NLP(Natural Language Processing). In many previous research a HMM(Hidden Markov Model) with unknown parameters has been used as a language model. In this research we also use the HMM as the language model. And we assume that sate transitions in the HMM are dominated by a second order Markov chain. At first we propose two types of morphological analysis methods which minimize the error rate with reference to a Bayes criterion. But the computational complexity of the proposed Bayes optimal morphological analysis methods are exponential order. So we also propose approximate methods.",
    keywords = "Hidden markov model, Morphological analysis, Statistical decision theory",
    author = "Yasunari Maeda and Naoya Ikeda and Hideki Yoshida and Yoshitaka Fujiwara and Toshiyasu Matsushima",
    year = "2007",
    doi = "10.1109/SICE.2007.4421232",
    language = "English",
    isbn = "4907764286",
    pages = "1563--1568",
    booktitle = "Proceedings of the SICE Annual Conference",

    }

    TY - GEN

    T1 - A note on morphological analysis methods based on statistical decision theory

    AU - Maeda, Yasunari

    AU - Ikeda, Naoya

    AU - Yoshida, Hideki

    AU - Fujiwara, Yoshitaka

    AU - Matsushima, Toshiyasu

    PY - 2007

    Y1 - 2007

    N2 - Morphological analysis is one of important topics in the field of NLP(Natural Language Processing). In many previous research a HMM(Hidden Markov Model) with unknown parameters has been used as a language model. In this research we also use the HMM as the language model. And we assume that sate transitions in the HMM are dominated by a second order Markov chain. At first we propose two types of morphological analysis methods which minimize the error rate with reference to a Bayes criterion. But the computational complexity of the proposed Bayes optimal morphological analysis methods are exponential order. So we also propose approximate methods.

    AB - Morphological analysis is one of important topics in the field of NLP(Natural Language Processing). In many previous research a HMM(Hidden Markov Model) with unknown parameters has been used as a language model. In this research we also use the HMM as the language model. And we assume that sate transitions in the HMM are dominated by a second order Markov chain. At first we propose two types of morphological analysis methods which minimize the error rate with reference to a Bayes criterion. But the computational complexity of the proposed Bayes optimal morphological analysis methods are exponential order. So we also propose approximate methods.

    KW - Hidden markov model

    KW - Morphological analysis

    KW - Statistical decision theory

    UR - http://www.scopus.com/inward/record.url?scp=50249153088&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=50249153088&partnerID=8YFLogxK

    U2 - 10.1109/SICE.2007.4421232

    DO - 10.1109/SICE.2007.4421232

    M3 - Conference contribution

    AN - SCOPUS:50249153088

    SN - 4907764286

    SN - 9784907764289

    SP - 1563

    EP - 1568

    BT - Proceedings of the SICE Annual Conference

    ER -