摘要 |
The present invention relates to a solution configured to analyze an emotion and a tendency of a certain person, by analyzing a language used by the person and, more specifically, to a solution configured to analyze an emotion and a tendency of a person who write or speaks (hereinafter referred to as an object to be analyzed) by analyzing a corpus (hereinafter referred to as a text). The solution comprises: a step of acquiring a text file; a step of classifying a sentence as a vocabulary unit; a step of converting a vocabulary to a morpheme unit to extract a morpheme; a step of classifying the extracted morpheme into a substantive group and a predicate group; a step of providing an identification tag to the extracted morpheme by classifying the extracted morpheme into any one among a native language, a Chinese language, a foreign language, and a loan word; a step of providing an identification tag to the extracted morpheme depending on each pre-determined part of speech; a step of filtering an emotion and a tendency depending on a pre-configured substantive category; a step of filtering an emotion and a tendency depending on a pre-configured predicate category; a step of calculating an emotion value and a tendency value of a relevant substantive category; and a step of calculating an emotion value and a tendency value of a relevant predicate category. |