METHOD FOR EXTRACTING MAJOR SEMANTIC FEATURE FROM VOICE OF CUSTOMER DATA AND DATA CONCEPT CLASSIFICATION METHOD USING THEREOF

摘要

The present invention relates to a method for extracting a main feature from voice-of-customer (VOC) data and a data type classification method using the same. According to an embodiment of the present invention, the method for extracting a main feature from voice-of-customer (VOC) data by using a lexico-semantic pattern (LSP) and a data type classification method using the same can comprise: (a) a step of defining a type, a semantic feature, and an LSP for VOC data to construct LSP knowledge in advance; (b) a step of selecting a semantic feature candidate group from the constructed LSP knowledge; (c) a step of extracting a new semantic feature for the VOC data based on the constructed LSP knowledge and the selected semantic feature candidate group; and (d) a step of classifying a type for newly inputted VOC data based on the extracted new semantic feature and the constructed LSP knowledge. According to an embodiment of the present invention, the type of newly inputted voice-of-customer (VOC) data can be easily classified and information of main features for the VOC data can be acquired.