A new method for selecting English field association terms of compound words and its knowledge representation |
| |
| Authors: | El-Sayed Atlam K Morita M Fuketa Jun-ichi Aoe |
| |
| Institution: | 1. Technical Research Institute, SAMMI Information System, Seoul 08502, Republic of Korea;2. Graduate School of Information Science and Engineering, Ritsumeikan University, Shiga 5258577, Japan;3. Department of electronic engineering, Inha University, Incheon 402-751, Republic of Korea;1. Department of Technology and Innovation, University of Southern Denmark, Odense, Denmark;2. Department of Green Technology, University of Southern Denmark, Odense, Denmark |
| |
| Abstract: | This paper presents a strategy for building a morphological machine dictionary of English that infers meaning of derivations by considering morphological affixes and their semantic classification. Derivations are grouped into a frame that is accessible to semantic stem and knowledge base. This paper also proposes an efficient method for selecting compound Field Association (FA) terms from a large pool of single FA terms for some specialized fields. For single FA terms, five levels of association are defined and two ranks are defined, based on stability and inheritance. About 85% of redundant compound FA terms can be removed effectively by using levels and ranks proposed in this paper. Recall averages of 60–80% are achieved, depending on the type of text. The proposed methods are applied to 22,000 relationships between verbs and nouns extracted from the large tagged corpus. |
| |
| Keywords: | Compound terms Field association term Semantic classification Redundant candidates |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
