, Le graphique (a) représente l'évolution du MAE, sur les ensembles de test, en fonction de la taille du corpus d'apprentissage. Le graphique (b) représente la même évolution, Figure 5.16 -L'influence du nombre de données pour le corpus d'apprentissage équilibré

, En équili-brant les données d'apprentissage, la fonction de prédiction perd en qualité. En effet, l'erreur de prédiction, estimée par le score MAE, dépasse en moyenne le 0,125, comparée à une erreur inférieure à 0,125 avec l'ensemble déséquilibré. Sur l'ensemble de test déséquilibré, la fonction de prédiction s'améliore en augmentant la taille du corpus d'apprentissage, tandis que la variation n'influence pas les performances sur l, Dans la figure 5.16, nous présentons les mêmes expériences, sauf que dans ce cas nous évaluons les performances du système appris en utilisant le corpus d'apprentissage équilibré

, Pour les données équilibrées, le score EAM est autour de 0,3 pour les traductions parfaites (BLEU=1), tandis qu'il est de 0,35 en moyenne pour les données d'apprentissage déséquilibrées. Pour les traductions ayant un score BLEU entre 0,25 et 0,5, les plus nombreuses dans le corpus de test (voir Figure 5.11), le score EAM est inférieure à 0,1 en moyenne pour les données d'apprentissage déséquilibrées, La dégradation de la qualité de la fonction de prédiction en utilisant un corpus d'apprentissage équilibré, s'explique par l'analyse du graphique (b) de la figure 5.16, vol.5

. Moses-nmt, , vol.33, p.8

. Moses-nmt-gamat, , vol.34, p.38

, 13 -Comparaison et combinaison oracle des trois systèmes au niveau de la phrase en utilisant la métrique BLEU, vol.5

G. De-manière, Grâce à ces deux cas de combinaison, en plus d'améliorer le score BLEU global sur l'ensemble de test, GAMaT parvient à faire mieux que MOSES dans 31.12% des cas. Il obtient les mêmes performances que MOSES dans 19.26% des cas. Comparé à NMT, il obtient de meilleures performances dans 33,02% des cas, et fait de même dans 7,42% des cas. La combinaison de MOSES avec NMT permet d'obtenir une performance globale de 33,77 en BLEU, ce qui est légèrement supérieur à la combinaison entre NMT et GAMaT (33,06 de BLEU), nous observons que toutes les combinaisons du Tableau 5.13, permettent une importante amélioration des performances de traduction des systèmes combinés

;. La-dernière-combinaison and . Moses-nmt-gamat, consiste à combiner les trois systèmes en même temps. Nous observons que cette combinaison permet d'obtenir les meilleures performances de traduction sur l'ensemble de test. En effet, vol.50, pp.69-2707

. Moses-nmt, , vol.50, p.6

. Moses-nmt-gamat, , vol.49, p.4

. Al-onaizan, Statistical Machine Translation, Final Report, JHU Summer Workshop, vol.30, 1999.

L. Araujo-;-araujo, How Evolutionary Algorithms are Applied to Statistical Natural Language Processing, Artificial Intelligence Review, vol.28, issue.4, pp.275-303, 2007.

A. , Machine Translation : an Introductory Guide. Blackwells-NCC, 1993.

T. Back-;-back, Evolutionary Algorithms in Theory and Practice : Evolution Strategies, Evolutionary Programming, Genetic Algorithms, 1996.

[. Bahdanau, METEOR : an Automatic Metric for MT Evaluation with Improved Correlation with Human Judgments, Proceedings of the ACL Workshop on Intrinsic and Extrinsic Evaluation Measures for Machine Translation and/or Summarization, pp.65-72, 2005.

H. Beyer and H. Schwefel, Evolution Strategies-A Comprehensive Introduction, Natural Computing, vol.1, issue.1, pp.3-52, 2002.

S. Binitha, S. Binitha, and S. S. Sathya, A Survey of Bio Inspired Optimization Algorithms, International Journal of Soft Computing and Engineering, issue.2, 2012.

. Boitet and C. Boitet, Pros and Cons of the Pivot and Transfer Approaches in Multilingual Machine Translation. Readings in Machine Translation, pp.273-279, 1988.

[. Bojar, Proceedings of the Second Conference on Machine Translation, 2017.

[. Bojar, Proceedings of the Tenth Workshop on Statistical Machine Translation, pp.1-46, 2015.

[. Bonabeau, Swarm Intelligence : from Natural to Artificial Systems. Number 1, 1999.

R. Bossard, A. Bossard, and C. Rodrigues, Une Approche Évolu-tionnaire pour le Résumé Automatique, TALN 2015-22ème Conférence sur le Traitement Automatique des Langues Naturelles, vol.22, 2015.

[. Brown, A Statistical Approach to Machine Translation, Computational linguistics, vol.16, issue.2, pp.79-85, 1990.

[. Brown, Aligning Sentences in Parallel Corpora, Proceedings of the 29th Annual Meeting on Association for Computational Linguistics, ACL '91, pp.169-176, 1991.

[. Brown, The Mathematics of Statistical Machine Translation : Parameter Estimation, Comput. Linguist, vol.19, issue.2, pp.263-311, 1993.

. Callison-burch, Findings of the 2010 Joint Workshop on Statistical Machine Translation and Metrics for Machine Translation, Proceedings of the Joint Fifth Workshop on Statistical Machine Translation and MetricsMATR, pp.17-53, 2010.

Z. Chen, H. Chen, and Y. Zhu, Optimization Based on Symbiotic Multispecies Coevolution, Applied Mathematics and Computation, vol.205, issue.1, pp.47-60, 2008.

G. Chen, S. F. Chen, and J. Goodman, An Empirical Study of Smoothing Techniques for Language Modeling, Computer Speech & Language, vol.13, issue.4, pp.359-394, 1999.

[. Chiang, Two Easy Improvements to Lexical Weighting, Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics : Human Language Technologies : short papers, vol.2, pp.455-460, 2011.

C. , Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling, 2014.

[. ?repin?ek, Exploration and Exploitation in Evolutionary Algorithms : a Survey, ACM Computing Surveys (CSUR), vol.45, issue.3, p.35, 2013.

. Devlin, Fast and Robust Neural Network Joint Models for Statistical Machine Translation, Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, vol.1, pp.1370-1380, 2014.

[. Dorigo, Ant System : Optimization by a Colony of Cooperating Agents, IEEE Transactions on Systems, Man, and Cybernetics, vol.26, issue.1, pp.29-41, 1996.

[. Douib, Genetic-based Decoder for Statistical Machine Translation, 2016.
URL : https://hal.archives-ouvertes.fr/hal-01336546

[. Douib, GAWO : Genetic-based Optimization Algorithm for SMT, ICNLSSP 2017-International Conference on Natural Language, Signal and Speech Processing, 2017.
URL : https://hal.archives-ouvertes.fr/hal-01660010

[. Douib, A Translation Evaluation Function Based on Neural Network, Schedae Informaticae, vol.25, pp.139-151, 2016.
URL : https://hal.archives-ouvertes.fr/hal-01531584

. Echizen-ya, Machine Translation Method Using Inductive Learning with Genetic Algorithms, Proceedings of the 16th Conference on Computational Linguistics, vol.2, pp.1020-1023, 1996.

J. L. Elman, Finding Structure in Time, Cognitive Science, vol.14, issue.2, pp.179-211, 1990.

D. B. Fogel, Foundations of Evolutionary Computation, Modeling and Simulation for Military Applications, vol.6228, p.622801, 2006.

F. Glover-;-glover, Tabu Search-Part I, ORSA Journal on Computing, vol.1, issue.3, pp.190-206, 1989.

J. González, A Finite-state Approach to Phrase-based Statistical Machine Translation, Proceedings of the 10th International Workshop on Finite State Methods and Natural Language Processing, pp.90-98, 2012.

W. Han, A. L. Han, and D. F. Wong, Machine Translation Evaluation : a Survey, 2016.

B. Han-;-han, Building a Bilingual Dictionary with Scarce Resources : a Genetic Algorithm Approach, Proceedings of the Student Research Workshop, the Second Meeting of the North American Chapter, 2001.

[. Hasler, Margin Infused Relaxed Algorithm for MOSES, The Prague Bulletin of Mathematical Linguistics, vol.96, pp.69-78, 2011.

S. Hochreiter and J. Schmidhuber, Long Shortterm Memory, Neural Computation, vol.9, issue.8, pp.1735-1780, 1997.

J. H. Holland-;-holland, Genetic Algorithms and the Optimal Allocation of Trials, SIAM Journal on Computing, vol.2, issue.2, pp.88-105, 1973.

M. Hopkins and J. May, Tuning as Ranking, Proceedings of the Conference on Empirical Methods in Natural Language Processing, pp.1352-1362, 2011.

W. J. Hutchins, Machine Translation : a Brief History, Concise History of the Language Sciences, pp.431-445, 1995.

W. J. Hutchins-;-hutchins, Machine Translation over Fifty Years. Histoire, Epistemologie, vol.23, pp.7-31, 2001.

X. Hüe-;-hüe, Genetic Algorithms for Optimisation : Background and Applications, 1997.

S. Katz-;-katz, Estimation of Probabilities from Sparse Data for the Language Model Component of a Speech Recognizer, IEEE Transactions on Acoustics, Speech, and Signal Processing, vol.35, issue.3, pp.400-401, 1987.

E. Kennedy, J. Kennedy, and R. Eberhart, Particle swarm optimization, Proceedings of ICNN'95 -International Conference on Neural Networks, vol.4, pp.1942-1948, 1995.

[. Klein, Opennmt : Open-source Toolkit for Neural Machine Translation, 2017.

. Kocur, V. Bojar-;-kocur, and O. Bojar, Particle Swarm Optimization Submission for WMT16 Tuning Task, Proceedings of the First Conference on Machine Translation, pp.518-524, 2016.

P. Koehn-;-koehn, Pharaoh : a Beam Search Decoder for Phrase-based Statistical Machine Translation Models. Machine translation : From Real Users to Research, pp.115-124, 2004.

[. Koehn, MOSES : Open Source Toolkit for Statistical Machine Translation, Proceedings of the 45th Annual Meeting of the ACL on Interactive Poster and Demonstration Sessions, pp.177-180, 2007.

P. Koehn, R. Knowles, and . Koehn, Statistical Phrase-based Translation, Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology, vol.1, pp.48-54, 2003.

J. R. Koza, Genetic Programming as a Means for Programming Computers by Natural Selection, Statistics and Computing, vol.4, issue.2, pp.87-112, 1994.

. Langlais, A Greedy Decoder for Phrasebased Statistical Machine Translation, Proc. of TMI, 2007.

D. Langlois, Loria System for the WMT15 Quality Estimation Shared Task, Proceedings of the Tenth Workshop on Statistical Machine Translation, pp.323-329, 2015.
URL : https://hal.archives-ouvertes.fr/hal-01262104

, A Survey on the Use of Genetic Algorithms in Natural Language Processing, Proceedings of the 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics, vol.1, pp.1-5, 1998.

A. R. Rafea-;-nabhan, A. Rafea, M. Nagao, G. Neubig, and T. Watanabe, A Framework of a Mechanical Translation between Japanese and English by Analogy Principle, Information Reuse and Integration, Conf, 2005. IRI-2005 IEEE International Conference on, pp.1-54, 1984.

F. J. Och, An Efficient Method for Determining Bilingual Word Classes, Proceedings of the Ninth Conference on European Chapter of the Association for Computational Linguistics, pp.71-76, 1999.

F. J. Och, Statistical Machine Translation : from Single Word Models to Alignment Templates, 2002.

F. J. Och, Minimum Error Rate Training in Statistical Machine Translation, Proceedings of the 41st Annual Meeting on Association for Computational Linguistics, vol.1, pp.160-167, 2003.

F. J. Ney-;-och and H. Ney, A Systematic Comparison of Various Statistical Alignment Models, Comput. Linguist, vol.29, issue.1, pp.19-51, 2003.

R. Otto, E. Otto, M. C. Riff, C. H. Papadimitriou, K. Steiglitz et al., Towards an Efficient Evolutionary Decoding Algorithm for Statistical Machine Translation, Proceedings of the 40th Annual Meeting on Association for Computational Linguistics, pp.311-318, 1998.

C. Ross, P. Corne, and D. , Applications of Genetic Algorithms, AISB Quaterly on Evolutionary Computation, vol.89, pp.23-30, 1994.

[. Snover, A Study of Translation Edit Rate with Targeted Human Annotation, Proceedings of Association for Machine Translation in the Americas, vol.200, 2006.

, SRILM at Sixteen : Update and Outlook, Proceedings of IEEE Automatic Speech Recognition and Understanding Workshop, vol.5, 2011.

, A New Quantitative Quality Measure for Machine Translation Systems, Proceedings of the 14th Conference on Computational Linguistics, vol.2, pp.28-36, 1985.

M. Willmott, C. Willmott, and K. Matsuura, Advantages of the Mean Absolute Error (MAE) Over the Root Mean Square Error (RMSE) in Assessing Average Model Performance. Climate research, vol.30, pp.79-82, 2005.

. Zens, Efficient Conflict Driven Learning in a Boolean Satisfiability Solver, Annual Conference on Artificial Intelligence, pp.18-32, 2001.

A. Zogheib, Genetic Algorithm-based Multi-word Automatic Language Translation. Recent Advances in Intelligent Information Systems, IEEE/ACM International Conference on Computer-aided Design, pp.751-760, 2011.