WEB INFORMATION COLLECTION DEVICE, WEB CRAWLER PROGRAM AND WEB INFORMATION COLLECTION METHOD

摘要

PPROBLEM TO BE SOLVED: To provide a Web information collection device, a Web crawler program and a Web information collection method, not requiring manual production and update of learning data, automatically collecting only a Web page that is a target, and not collecting information that is noise to the utmost. PSOLUTION: This Web information collection device 2 has a control part 4 cyclically acquiring Web information from Web servers 3a, 3b, 3c, deciding whether the Web information is target information related to a specific field or non-target information, acquiring second Web information about a link source to extract a word included in a text information portion, learning cumulative frequency of each word in each case of the target information or the non-target information, calculating priority by appearance frequency of the extracted word and the learned cumulative frequency, and determining a link destination preferentially accessed by largeness of the priority, and collects the Web information while tracing link information. PCOPYRIGHT: (C)2006,JPO&NCIPI