医学知識発見におけるルールの興味深さ指標の評価  [in Japanese] Evaluation of Rule Interestingness Measures in Medical Knowledge Discovery in Databases  [in Japanese]

Access this Article

Search this Article

Author(s)

Abstract

本論文では,医療データセットを用いた2 つの実証実験を通して,医学知識発見における興味深さ指標の有用性を調べ,興味深さ指標の活用方法を議論する.まず,髄膜炎の医療データセットから生成されたルールに対し,40 種類の多様な興味深さ指標と医師による評価結果を得た.そして,F-Measureと相関係数を使って興味深さ指標が持つ医師の興味の推定性能を見積もり,興味深さ指標間で比較した.肝炎の医療データセットでも同様の実験を行った.その結果,Accuracy,Uncovered Negative,Peculiarity,Relative Risk,Chi-Square Measure for One Quadran は,医学ドメインでの人間の興味を安定かつ妥当に推定する可能性が示された.また,推定性能は医師が立てる仮説の確かさに依存すること,興味深さ指標を組み合わせて,マイニングシステムとのインタラクションを通した医師の仮説生成・検証を支援しうることが示唆された.We discuss the usefulness of rule interestingness measures for medical KDD through experiments using clinical datasets and also consider how to utilize these measures in postprocessing based on the experimental outcomes. We first conducted an experiment to compare evaluation results by forty various interestingness measures with those by a medical expert for rules discovered in a clinical dataset on meningitis. We calculated and compared the performance of each interestingness measure to estimate a medical expert's interest with the F-Measure and correlation coefficient. We then conducted a similar experiment on hepatitis. The comprehensive results of experiments on meningitis and hepatitis showed that Accuracy, Uncovered Negative, Peculiarity, Relative Risk, and Chi-Square Measure for One Quadrant have a stable, reasonable performance in estimating real human interest in the medical domain. The results also indicate that the performance of interestingness measures is influenced by the certainty of a hypothesis made by a medical expert, and that the combinational use of interestingness measures will contribute to support medical experts to generate and confirm their hypotheses through human-system interaction.

We discuss the usefulness of rule interestingness measures for medical KDD through experiments using clinical datasets and also consider how to utilize these measures in postprocessing based on the experimental outcomes. We first conducted an experiment to compare evaluation results by forty various interestingness measures with those by a medical expert for rules discovered in a clinical dataset on meningitis. We calculated and compared the performance of each interestingness measure to estimate a medical expert's interest with the F-Measure and correlation coefficient. We then conducted a similar experiment on hepatitis. The comprehensive results of experiments on meningitis and hepatitis showed that Accuracy, Uncovered Negative, Peculiarity, Relative Risk, and Chi-Square Measure for One Quadrant have a stable, reasonable performance in estimating real human interest in the medical domain. The results also indicate that the performance of interestingness measures is influenced by the certainty of a hypothesis made by a medical expert, and that the combinational use of interestingness measures will contribute to support medical experts to generate and confirm their hypotheses through human-system interaction.

Journal

  • IPSJ journal

    IPSJ journal 48(4), 1859-1873, 2007-04-15

    Information Processing Society of Japan (IPSJ)

References:  45

Cited by:  1

Codes

  • NII Article ID (NAID)
    110006251458
  • NII NACSIS-CAT ID (NCID)
    AN00116647
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    1882-7764
  • NDL Article ID
    8785105
  • NDL Source Classification
    ZM13(科学技術--科学技術一般--データ処理・計算機)
  • NDL Call No.
    Z14-741
  • Data Source
    CJP  CJPref  NDL  NII-ELS  IPSJ 
Page Top