Pattern extracting apparatus and method

摘要

A pattern extracting apparatus includes a first storing section storing plural target information pieces, a candidate pattern producing section producing candidate patterns each including two or more items different from each other based on each of the items included in the plural target information pieces, a candidate evaluation value calculating section calculating an extraction evaluation value of the candidate pattern based on a frequency of appearance at which the produced candidate pattern appears in the plural target information pieces, a pattern extracting section determining and extracting any of the candidate patterns having the calculated extraction evaluation value satisfying a predetermined threshold value, and a second storing section storing an association degree between the items. The candidate evaluation value calculating section calculates the extraction evaluation value based on a weight based on an identified association degree between the items included in the candidate pattern and the frequency of appearance.

主权项

1. A pattern extracting apparatus extracting, from a plurality of items included in target information pieces, a pattern of a combination of two or more of the items different from each other, comprising:
a first storing section storing the plurality of target information pieces; a candidate pattern producing section producing candidate patterns each formed of two or more of the items different from each other based on each of the items included in each of the plurality of target information pieces; a candidate evaluation value calculating section calculating an extraction evaluation value of the candidate pattern; a pattern extracting section determining any of the candidate patterns having the calculated extraction evaluation value satisfying a predetermined threshold value and extracting the candidate pattern satisfying the threshold value; a second storing section storing an association degree between the items, wherein the candidate evaluation value calculating section extracts the association degree between the items included in the candidate pattern from the second storing section, calculates an association degree of the candidate pattern based on the extracted association degree between the items and a frequency of appearance at which the candidate pattern appears in each of the plurality of target information pieces, simultaneously evaluates the association degree of the candidate pattern and the frequency of appearance of the candidate pattern, and calculates the extraction evaluation value, wherein the candidate evaluation value calculating section calculates the extraction evaluation value using the calculated frequency of appearance and the evaluated association degree of the candidate pattern, such that the calculated extraction evaluation value is lower as the calculated frequency of appearance is lower and as the evaluated association degree of the candidate pattern is higher, and wherein the candidate evaluation value calculating section calculates the extraction evaluation value of the candidate pattern such that the calculated extraction evaluation value monotonously decreases as the number of the items constituting the candidate pattern increases, based on a combined degree of the candidate pattern for each association degree between the items constituting the candidate pattern in accordance with a number of combinations of the items; and provides for analysis the pattern of the combination of the two or more of the items different from each other when the calculated extraction evaluation value satisfies the predetermined threshold value.