Information Extraction from Electronic Mail
スポンサーリンク
概要
- 論文の詳細を見る
The former researchers on information extraction are only from documents with only sentences. But many documents include non-sentence areas, i.e., tabular areas such as itemizations, tabular forms and marsharing of nouns. This paper describes an information extraction method from tabular areas. The new alforithm consists of three steps. First, tabular areas are recognized by using a 2-dimensional arrangement of letters, blank spaces, parts of speech, and semantic features of nouns. Second, the tabular areas are subdivided into blocks. And last, information extraction is done by matching the block against the frame. Words themselves and semantic feature of words are the clues clues to fill in the slots in the frame. Desk simulation for electronic mail of computer sale information shows a precision rate of 83%, 92% and 78% for each step.
- Faculty of Engineering, Mie Universityの論文
- 1994-12-21
Faculty of Engineering, Mie University | 論文
- Approximate Method of Elastic Buckling Strength Analysis for Irregular Frames
- Elastic Buckling Strength of Steel Frames Connected to Core Wall
- Calculations of Energy Shift of the Conduction Band-Edge in Doped and Compensated GaP
- Estimation of the Glass Transition Pressure of Liquids under Very High Pressure by Viscoelastic Analysis
- Information Extraction from Electronic Mail