Hitachi, Ltd.

Yukiyasu Ito

Akira Maeda

Hitoshi Ashida

Toyohisa Morita

Yoji Taniguchi

Erika Ayukawa

Yori Takahashi

Yoshinori Sato

Hideyuki Maki

A reasoning method for determining similar cases by executing a case-similarity calculation step and utilizing a distribution of values of fields for calculation of reasoning results of cases having high similarities. Since the similar cases are determined taking a peripheral distribution of a new case into consideration, suitable similar cases can be selected for reasoning even when the case distribution is not uniform.

Reasoning method based on similarity of cases

Data analysis using online data summarization processing and rule-based data characteristic detection is performed at the time the transaction data is input. Online data input through an input device of a first server is stored into a transaction data memory device. A data summarizing device carries out summarization processing of online data according to stored parameters, and the resultant summary data is stored into a summary data memory device. A data joining device joins the summary data stored in the summary data memory device and the summary data transferred from a second server for the purpose of updating. In the first server, using rules generated by a rule generating device and stored in rule memory device, a detecting device detects characteristics of the summary data. Further, the results of data characteristic detection are sent to a client through.

Data analysis method and apparatus for data mining

A data analyzing method for generating a rule based on data items in a data base, wherein the rule expresses relational features of the data items. The invention includes a user interface and a rule generation module. The rule generation module, in response to an input from the user via the user interface, selects data items for use in a conditional clause and a conclusion clause of a rule from the data items stored in the data base, converts, when the selected data items have numerical values, the numerical values into symbolic values and creates plural candidate rules each expressing a correlation between selected data items in a rule form having one or plural sets of item names and symbolic values. The rule generation module further calculates a criterion for evaluating strength of correlation between data items in each of the candidate rules, determines one or plural candidate rules having highest calculated criterion from the candidate rules, and outputs to the user via the user interface the one or plural candidate rules.

Data analyzing method for generating rules

A data analyzing method and system for generating a rule based on data items in a data base, wherein the rule expresses relational features of the data items. The invention includes a user interface and a rule generation module. The rule generation module, in response to an input from the user via the user interface, selects data items for use in a conditional clause and a conclusion clause of a rule from the data items stored in the data base, converts, when the selected data items have numerical values, the numerical values into symbolic values and creates plural candidate rules each expressing a correlation between selected data items in a rule form having one or plural sets of item names and symbolic values. The rule generation module further calculates a criterion for evaluating strength of correlation between data items in each of the candidate rules, determines one or plural candidate rules having highest calculated criterion from the candidate rules, and outputs to the user via the user interface the one or plural candidate rules.

Data analyzing method

In a data utilizing system, a statistic table exhibiting a characteristic distribution is automatically extracted from given statistic tables of data. After confirmation of distribution of the data on the extracted statistic table, data required is retrieved from the statistic table. The data utilizing system for extracting from data constituted by a set of records each composed of at least two item values each containing at least one numeric item value, a part of the records constituting a part of the data, includes: as statistic table list generating unit for generating from the data statistic tables of component data for which at least one item value is definitely specified, and for determining arithmetically characteristic quantities for the statistic tables, respectively, on the basis of differences between the statistic tables and relevant expected value distribution tables, to thereby generate a list of the statistic tables on the basis of the characteristic quantities; and a record list generating unit for generating from the data a list of records contained in all or at least one specified cell of a specified given one of the statistic tables.

Growing community of inventors

Fujisawa, Japan

Yukiyasu Ito