| 摘要 |
PURPOSE: A content classification method based on morpheme and an apparatus thereof are provided to automatically map a content onto a category with high accuracy, thereby effectively classifying the content in real time. CONSTITUTION: A content collection unit (410) collects contents which are provided on the internet, and a morpheme extraction unit (420) obtains a morpheme by analyzing the title of the contents and identifies the location of the morpheme which is obtained from text information included in the title of the contents. A synonym information storage unit (430) stores a synonym dictionary and searches for a synonym corresponding to the morpheme. A weighting setup unit (440) gives a weighting to the obtained morpheme depending on the location of the morpheme. A category storage unit (450) classifies the contents into a sub category of the category by searching a category which is matched to the morpheme whose weighting has been determined. [Reference numerals] (AA) No; (BB) Yes; (S410) Search morpheme synonym; (S420) Search all morpheme synonym?; (S430) Synonym numbering |
