Name
Abstract | ||
|---|---|---|
Terminology mining is a major step forward in terminology extraction and covers acquisition and structuring of the candidate
terms. We presents a terminology mining method based on linguistic criteria and combined computational methods. In terminology
mining, references are made to the acquisition of complex terms, the discovering of new terms, but also, the structuring of
the acquired candidate terms. First, the linguistic specifications of terms are given for French and we define a typology
of base-terms and their variations. We stress the crucial part of the handling of term variations to build a linguistic structuring,
to detect advanced lexicalisation and to obtain an optimised representativity of the candidate term occurrences. Second, we
move to the computational methods implemented: shallow parsing, morphological analysis, morphological rule learning and lexical
statistics. Third, the system that identifies base terms and their variations, ACABIT (Automatic Corpus-Based Acquisition
of Binary Terms) is introduced: its architecture, the languages it applies on and its functions. To conclude, a review of
evaluation methods for terminology extraction is presented and results of the efficiency of ACABIT in evaluation campaigns
are discussed.
|
Year | DOI | Venue |
|---|---|---|
2002 | 10.1007/978-3-540-45092-4_2 | SCIE |
DocType | Citations | PageRank |
Conference | 3 | 0.44 |
References | Authors | |
11 | 1 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Béatrice Daille | 1 | 306 | 34.40 |