| 主权项 |
1. An information processing apparatus comprising:
a web page acquiring unit that acquires a plurality of web pages of an identical category into which a target item described in the plurality of web pages are classified; an attribute extracting unit, implemented by a processor, that extracts an attribute-related term of the attribute matching an input attribute description pattern, from the plurality of web pages; an attribute description pattern extracting unit that extracts an attribute description pattern matching an input attribute-related term from the plurality of web pages; a data input unit that inputs an initial attribute description pattern in the attribute extracting unit, or inputs an initial attribute-related term in the attribute description pattern extracting unit; an attribute scoring unit that scores the attribute-related term; and an attribute selecting unit that ranks the attribute-related term in order of the score, and selects an attribute-related term of a predetermined rank or more, wherein, the data input unit further inputs, when the attribute extracting unit extracts the attribute-related term, the extracted attribute-related term in the attribute description pattern extracting unit, or further inputs, when the description pattern extracting unit extracts the description pattern, the extracted description pattern in the attribute extracting unit, and wherein a website includes a plurality of stores which sell the target item, each of the stores having a web page in the website, wherein the attribute scoring unit scores the attribute-related term by counting, as a first count, a number of stores, among the plurality of stores, whose web pages include the attribute-related term, and wherein the attribute-related term having the first count, which is higher than a second count corresponding to another attribute-related term, is given a higher score than the other attribute-related term. |
