%0 Book Section %1 year=2003 %A Daille, Béatrice %B Information Extraction in the Web Era %D 2003 %E Pazienza, MariaTeresa %I Springer Berlin Heidelberg %K mining terminology %P 29-44 %R 10.1007/978-3-540-45092-4_2 %T Terminology Mining %U http://dx.doi.org/10.1007/978-3-540-45092-4_2 %V 2700 %X 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. %@ 978-3-540-40579-5

@incollection{year={2003}, 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.}, added-at = {2013-05-18T19:33:33.000+0200}, author = {Daille, Béatrice}, biburl = {https://www.bibsonomy.org/bibtex/2b3216103e76221fcc76c4fb3466eabdf/lovebooks}, booktitle = {Information Extraction in the Web Era}, doi = {10.1007/978-3-540-45092-4_2}, editor = {Pazienza, MariaTeresa}, interhash = {31ddb29497a5134cbb367153998a8837}, intrahash = {b3216103e76221fcc76c4fb3466eabdf}, isbn = {978-3-540-40579-5}, keywords = {mining terminology}, pages = {29-44}, publisher = {Springer Berlin Heidelberg}, series = {Lecture Notes in Computer Science}, timestamp = {2013-05-18T19:33:33.000+0200}, title = {Terminology Mining}, url = {http://dx.doi.org/10.1007/978-3-540-45092-4_2}, volume = 2700, year = 2003 }