Knowledge Discovery by Inferences

通过推理发现知识

• 批准号: 10143104
• 负责人: SATO Taisuke
• 金额: $ 54.85万
• 依托单位: Tokyo Institute of Technology
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research on Priority Areas (A)
• 财政年份: 1998
• 资助国家: 日本
• 起止时间: 1998 至 2000
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-10143104/
• 关键词: knowledge discovery; inductive inference; abduction; ILP; 記号的統計モデリング; 脳機能画像; 拡張アブダクション; 高速パターンマッチング; 事例ベース; 節管理; 抽象値; 統計的記号モデリング; データマイニングアルゴリズム; H-map; 決定木; 反駁推論; 発見

From the fiscal year 1999 to 2001, we conducted research in knowledge discovery and developed inference various methods that can cope with uncertainty and complexities in the real data as follows.T. Sato developed a symbolic-statistical modeling language PRISM. Arimura et al. developed two types of fast text pattern matching algorithm. Tsukimoto proposed logical regression analysis. K.Satoh continued the analysis of minimal case base required for representing concepts. Sakama proposed non-monotonic inverse resolution.Imai et al. developed an algorithm for computing maximally specific hypotheses for ILP. Haraguchi proposed data abstraction for decision trees. Yamamoto reconstructed the theoretical base of ILP based on the logic of Suggestion.M. Sato et al. studied the theory of refutable/inductive inference. Shinohara proposed reducing data dimension by his H-map. Ohsawa showed that his KeyGraph approach is effective by examples.

从1999财年到2001财年，我们在知识发现方面进行了研究，并开发了各种方法来处理真实数据中的不确定性和复杂性，具体如下：T·佐藤开发了符号统计建模语言PRISM。AriMura等人。开发了两种快速的文本模式匹配算法。Tsukimoto提出了逻辑回归分析。K.Satoh继续分析表述概念所需的最小案例库。Sakama提出了非单调逆解.Imai等.提出了一种计算ILP的最大特定性假设的算法。Haraguchi提出了用于决策树的数据抽象。山本以Suggestion.M.Sato等人的逻辑为基础，重构了ILP的理论基础。研究了可反驳/归纳推理理论。Shinohara建议通过他的H-map来降低数据维度。Ohsawa通过实例证明了他的关键图法是有效的。

项目成果

Y.Mukouchi and M.Sato: "Language learning with a neighbor system"Proc.of DC 2000. LNAI1967. 183-196 (2000)

Y.Mukouchi 和 M.Sato：“使用邻居系统进行语言学习”Proc.of DC 2000。LNAI1967。

C.Sakama: "Inverse entailment in nonmonotonic logic programs"Proc.of ILP 2000. LNAI1866. 209-224 (2000)

C.Sakama：“非单调逻辑程序中的逆蕴涵”Proc.of ILP 2000。LNAI1866。

Y. KUDO, M. HARAGUCHI: "An Appropriate Abstraction for an Attribute-Oriented Induction"Proc. of DS '99. LNAI. 43-55 (1999)

Y. KUDO，M. HARAGUCHI：“面向属性归纳的适当抽象”Proc。

Koichi Furukawa: "On the Completion of the Most Specific Hypothesis Computation in Inverse Entailment for Mutual Recursion" Proc.DS'98,LNAI 1532,Springer-Verlag. (1998)

Koichi Furukawa：“关于互递归逆蕴涵中最具体假设计算的完成”Proc.DS98，LNAI 1532，Springer-Verlag。

A.Yamamoto.: "An Inference Method for the Complete Inverse of Relative Subsumption" New Generation Computing. 17. 99-117 (1999)

A.Yamamoto.：“相对包含的完全逆的推理方法”新一代计算。