According to an embodiment of the present invention, a method for inferring metadata of a creator by dividing text content into prescribed units comprises: a step of receiving text content; a step of dividing the text content into prescribed units to create one or more pieces of divided content; a step of dividing the one or more pieces of divided content into one or more morphemes, and determining a morpheme vector for the one or more pieces of divided content from the divided one or more morphemes based on a morpheme-vector converter; a step of dividing the one or more pieces of divided content into one or more syllables, and determining a syllables vector for the one or more pieces of divided content from the divided one or more syllables based on a syllables-vector converter; and a step of determining a metadata vector corresponding to metadata of a creator based on the morpheme vector for the one or more pieces of divided content, the syllable vector for the one or more pieces of divided content, and a creator recognizer.
展开▼