초록 ▼

A computer implemented system, two-step method and computer program product which improves the operations of multi-feature extraction and which efficiently develops classification rules from a large training database. Specifically, given a large training set of data tuples, the first phase, called t

A computer implemented system, two-step method and computer program product which improves the operations of multi-feature extraction and which efficiently develops classification rules from a large training database. Specifically, given a large training set of data tuples, the first phase, called the feature identification phase, identifies features, which have good power in separating data tuples, based on a subset of the training set. A feature that has a good power in correlating data tuples into groups is said to have a good discriminating power. Since the feature identification phase is performed on a subset of the training set, processing costs are minimized. Limiting this phase to the identification of features having good discriminating power ensures that the use of a subset of the training set does not adversely affect the validity of the conclusions drawn therefrom. In the second phase, called the feature combination phase, the identified features are evaluated in combination against the entire training set to determine the final classification rules. The prior identification of features having good discriminating power advantageously minimizes processing costs during this phase (which is run against the entire training set). The use of multiple features advantageously increases the discrimination power beyond that of the individual features.

대표청구항 ▼

[ We claim:] [1.] A computer-implemented method of mining classification rules from a computer database which includes a plurality of data tuples, each data tuple including a plurality of features, the database further including a training set including the data tuples, each data tuple in the traini

[ We claim:] [1.] A computer-implemented method of mining classification rules from a computer database which includes a plurality of data tuples, each data tuple including a plurality of features, the database further including a training set including the data tuples, each data tuple in the training set including a group identifier, comprising the steps of:(a) evaluating each feature in a subset of the training set as a function of its correlation with the group identifier, and identifying said features, responsive to said step of evaluating, having an ID score exceeding a predetermined ID threshold;(b) combining identified features into feature pairs and evaluating each feature pair in the training set as a function of its correlation with said any one of the group identifiers, and identifying said feature pairs having a combination score which exceeds a predetermined feature combination threshold; and(c) classifying the database as a function the identified features and the identified feature pair.