摘要 |
Provided are a method and device for modeling electronic documents. A method for modeling electronic documents by an electronic device includes: a frame acquisition step in which a first semantic frame and a second semantic frame are acquired from an electronic document; and a similarity determining step in which a content similarity value indicating similarity between arguments included in the first and second semantic frames is determined, a context similarity value indicating similarity between a predetermined number of frames adjacent to the first semantic frame and a predetermined number of frames adjacent to the second semantic frame is determined, and a frame similarity value between the first semantic frame and the second semantic frame is determined on the basis of the content similarity value and the context similarity value. |