Robustness of DNA-based clustering

Rohani Abu Bakar, Chu Yu-Yi, Junzo Watada

    Research output: Chapter in Book/Report/Conference proceedingChapter

    Abstract

    The primary objective of clustering is to discover a structure in the data by forming some number of clusters or groups. In order to achieve optimal clustering results in current soft computing approaches, two fundamental questions should be considered; (i) how many clusters should be actually presented in the given data, and (ii) how real or good the clustering itself is. Based on these two fundamental questions, almost clustering method needs to determine the number of clusters. Yet, it is difficult to determine an optimal number of a cluster group should be obtained for each data set. Hence, DNA-based clustering algorithms were proposed to solve clustering problem without considering any preliminary parameters such as a number of clusters, iteration and, etc. Because of the nature of processes between DNA-based solutions with a silicon-based solution, the evaluation of obtained results from DNA-based clustering is critical to be conducted. It is to ensure that the obtained results from this proposal can be accepted as well as other soft computing techniques. Thus, this study proposes two different techniques to evaluate the DNA-based clustering algorithms either it can be accepted as other soft computing techniques or the results that obtained from DNA-based clustering are not reliable for employed.

    Original languageEnglish
    Title of host publicationStudies in Computational Intelligence
    Pages75-92
    Number of pages18
    Volume372
    DOIs
    Publication statusPublished - 2011

    Publication series

    NameStudies in Computational Intelligence
    Volume372
    ISSN (Print)1860949X

    Fingerprint

    DNA
    Soft computing
    Clustering algorithms
    Silicon

    ASJC Scopus subject areas

    • Artificial Intelligence

    Cite this

    Bakar, R. A., Yu-Yi, C., & Watada, J. (2011). Robustness of DNA-based clustering. In Studies in Computational Intelligence (Vol. 372, pp. 75-92). (Studies in Computational Intelligence; Vol. 372). https://doi.org/10.1007/978-3-642-11739-8_4

    Robustness of DNA-based clustering. / Bakar, Rohani Abu; Yu-Yi, Chu; Watada, Junzo.

    Studies in Computational Intelligence. Vol. 372 2011. p. 75-92 (Studies in Computational Intelligence; Vol. 372).

    Research output: Chapter in Book/Report/Conference proceedingChapter

    Bakar, RA, Yu-Yi, C & Watada, J 2011, Robustness of DNA-based clustering. in Studies in Computational Intelligence. vol. 372, Studies in Computational Intelligence, vol. 372, pp. 75-92. https://doi.org/10.1007/978-3-642-11739-8_4
    Bakar RA, Yu-Yi C, Watada J. Robustness of DNA-based clustering. In Studies in Computational Intelligence. Vol. 372. 2011. p. 75-92. (Studies in Computational Intelligence). https://doi.org/10.1007/978-3-642-11739-8_4
    Bakar, Rohani Abu ; Yu-Yi, Chu ; Watada, Junzo. / Robustness of DNA-based clustering. Studies in Computational Intelligence. Vol. 372 2011. pp. 75-92 (Studies in Computational Intelligence).
    @inbook{f29cfbc09a874bcb820219de7e12c551,
    title = "Robustness of DNA-based clustering",
    abstract = "The primary objective of clustering is to discover a structure in the data by forming some number of clusters or groups. In order to achieve optimal clustering results in current soft computing approaches, two fundamental questions should be considered; (i) how many clusters should be actually presented in the given data, and (ii) how real or good the clustering itself is. Based on these two fundamental questions, almost clustering method needs to determine the number of clusters. Yet, it is difficult to determine an optimal number of a cluster group should be obtained for each data set. Hence, DNA-based clustering algorithms were proposed to solve clustering problem without considering any preliminary parameters such as a number of clusters, iteration and, etc. Because of the nature of processes between DNA-based solutions with a silicon-based solution, the evaluation of obtained results from DNA-based clustering is critical to be conducted. It is to ensure that the obtained results from this proposal can be accepted as well as other soft computing techniques. Thus, this study proposes two different techniques to evaluate the DNA-based clustering algorithms either it can be accepted as other soft computing techniques or the results that obtained from DNA-based clustering are not reliable for employed.",
    author = "Bakar, {Rohani Abu} and Chu Yu-Yi and Junzo Watada",
    year = "2011",
    doi = "10.1007/978-3-642-11739-8_4",
    language = "English",
    isbn = "9783642117381",
    volume = "372",
    series = "Studies in Computational Intelligence",
    pages = "75--92",
    booktitle = "Studies in Computational Intelligence",

    }

    TY - CHAP

    T1 - Robustness of DNA-based clustering

    AU - Bakar, Rohani Abu

    AU - Yu-Yi, Chu

    AU - Watada, Junzo

    PY - 2011

    Y1 - 2011

    N2 - The primary objective of clustering is to discover a structure in the data by forming some number of clusters or groups. In order to achieve optimal clustering results in current soft computing approaches, two fundamental questions should be considered; (i) how many clusters should be actually presented in the given data, and (ii) how real or good the clustering itself is. Based on these two fundamental questions, almost clustering method needs to determine the number of clusters. Yet, it is difficult to determine an optimal number of a cluster group should be obtained for each data set. Hence, DNA-based clustering algorithms were proposed to solve clustering problem without considering any preliminary parameters such as a number of clusters, iteration and, etc. Because of the nature of processes between DNA-based solutions with a silicon-based solution, the evaluation of obtained results from DNA-based clustering is critical to be conducted. It is to ensure that the obtained results from this proposal can be accepted as well as other soft computing techniques. Thus, this study proposes two different techniques to evaluate the DNA-based clustering algorithms either it can be accepted as other soft computing techniques or the results that obtained from DNA-based clustering are not reliable for employed.

    AB - The primary objective of clustering is to discover a structure in the data by forming some number of clusters or groups. In order to achieve optimal clustering results in current soft computing approaches, two fundamental questions should be considered; (i) how many clusters should be actually presented in the given data, and (ii) how real or good the clustering itself is. Based on these two fundamental questions, almost clustering method needs to determine the number of clusters. Yet, it is difficult to determine an optimal number of a cluster group should be obtained for each data set. Hence, DNA-based clustering algorithms were proposed to solve clustering problem without considering any preliminary parameters such as a number of clusters, iteration and, etc. Because of the nature of processes between DNA-based solutions with a silicon-based solution, the evaluation of obtained results from DNA-based clustering is critical to be conducted. It is to ensure that the obtained results from this proposal can be accepted as well as other soft computing techniques. Thus, this study proposes two different techniques to evaluate the DNA-based clustering algorithms either it can be accepted as other soft computing techniques or the results that obtained from DNA-based clustering are not reliable for employed.

    UR - http://www.scopus.com/inward/record.url?scp=80054719147&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=80054719147&partnerID=8YFLogxK

    U2 - 10.1007/978-3-642-11739-8_4

    DO - 10.1007/978-3-642-11739-8_4

    M3 - Chapter

    SN - 9783642117381

    VL - 372

    T3 - Studies in Computational Intelligence

    SP - 75

    EP - 92

    BT - Studies in Computational Intelligence

    ER -