Conclusion

Linguistic engineering techniques are gradually becoming an integral part of our working environment, as we seek to make better use of the information contained in electronic documents. The objectives of automatic language processing have diversified, as they respond to the now strategic need to access information contained in large volumes of text in an appropriate and relevant way. The original ambition of NLP to simulate language competence and perfectly reproduce the tasks of translators, proofreaders and documentalists has been radically transformed. Today, NLP systems must be robust enough to facilitate the activities of all types of users: these systems must be capable of processing real textual data, in noisy environments, to serve varied and precise application objectives (searching for texts, extracting information, classifying documents, etc.). This means translating complex objectives...