METHOD OF ANALYZING DATA FOR REPUTATION MANAGEMENT SYSTEM USING WEB CRAWLING

摘要

The present invention relates to a method of analyzing data for a reputation management system using web crawling. The purpose of the present invention is to construct a reputation management system through data analysis and data mining using web crawling. The method according to the present invention comprises the steps of: receiving a reputation management request for specific keywords from a client; allowing a data analysis apparatus to receive data that are web crawled using the specific keywords from a web server; separating the data into sentence units, and performing a language processing process with respect to each word within the separated sentences to extract linguistic qualities; calculating a negative opinion-determining index by referring to a negative morpheme dictionary per morphemes with respect to the extracted linguistic qualities, thereby calculating into a probability according to a weighted value an attribute priority score of the negative morphemes matching with the number of the linguistic qualities which match with negative morphemes in one sentence and exist within the one sentence, as a step of judging whether the specific keywords are a positive or negative opinion from the extracted linguistic qualities; judging that the specific keywords are a negative opinion if the index is a predetermined value or higher, and storing the specific keywords in a storage database (DB); aligning and classifying the specific keywords according to a domain address if the specific keywords are a negative opinion, and transmitting the aligned specific keywords to a reputation management server; and inputting results stored in the storage DB with respect to the specific keywords into the reputation management server to generate a report form.;COPYRIGHT KIPO 2017