A Connectionist Approach for the Automatic Processing of Syntactic Structures of the Arabic Language Based on Neo-Khalilian Formalism
Article Sidebar
Main Article Content
Abstract
The techniques using neural networks attempt to imitate the connectionist structure of the nervous system to extract the advantages of the capacities of lear-ning and generalisation (hardiness, tolerance to breakdowns, and the possibility of parallel treatment).
This survey enters in the setting of development of a connectionist system for the analysis of the syntactic structures of the arabic language based on the neo-khalilian linguistic formalism. It is undertaken in the multidisciplinary domain of the cognitive sciences, which integrates the mathematical modelisation by neural networks, the computer techniques of the automatic treatment of the natural language, and the basic concepts of a linguistic theory.
The conceived system untitled Neurokhall, is composed of a merely recurrent neurone network compound with a RAAM (Recursive Auto Associative Memory) for the recursive structure treatment. The system accepts in entrance the sentence word by word under the shape of syntactic features, and provides in exit the syn-tactic categories under the shape of the template of the level of the syntax of the neo-khalilian theory.
Article Details
References
__________, A Connectionist Parser with Recursive Sentence Structure and Lexical Disambiguation, AAAI – 92 Americain association for Artificial Intelligence, pp. 32-37, 1992.
Blair Alan D., Scalinhg-up RAAMs, Département informatique, université de Bandeis, janvier 1997.
Chafe Wallace, Meaning and the Structure of the Language, University Press Chicago, 1970.
Chan Samuel W.K. and Franklin James, A Neural Network Model for Acquisition of Semantic Structures, International Symposium on Speech Image Processing and Neural Networks, Hong Kong, pp. 221-224, 1994.
McClelland J.L., Mark st Jhon and Roman Taraban, « Sentence Comprehension : A Parallel Distributed Processing Approach », Language and Cognitive Processes, pp. 287- 335, 1989.
Elman Jeffrey L., « Finding Structure in Time », cognitive science, 14, pp. 179-211, 1990.
_____________, Distributed Representations, Simple Recurrent Networks, and Grammatical Structure, département des sciences cognitives et de linguistique, 1991.
Fodor.J and Pylyshyn.Z, « Connectionism and Cognitive Architecture : A Critical Analysis », in Connections and symbols, pp. 3-71, Cambridge MIT Press, 1988.
Hadj Salah Abderrahmane, Linguistique arabe et linguistique générale : Essai de méthodologie et d'épistémologie du ‘ilm al-‘Arabiyya, thèse de doctorat, Paris Sorbonne, Vol. 2, 1979.
Jain Jianchang Anil K. Mao K. Mohiuddin, Artificial Neural Networks : A Tutorial, IEEE Computer Special Issue on Neural Computing, mars 1996.
Koong H.C. Lin and Tung-Bo Chen and Von-Wun Soo, « Neural Network Learning and Encoding of Thematic Role Assignments in Parsing of Simple Chinese Sentences », Journal of Information Science and Engineering, vol.11, n° 1, pp.109-126,1995.
Meftouh Karima, Une approche connexionniste pour la génération d’une représentation interne du sens d’une phrase basée sur les cas sémantiques appliquée à la langue arabe, mémoire de Magistère, université de Annaba, 2000.
Miikkulainen Risto Michael G. Dyer, Natural Language Processing with Modular PDP Network and Distributed Lexicon, cognitive science, 1991.
Pollack Jordan B., Recursive Distributed Representation, Laboratoire de recherche en intelligence artificielle, université de l'état de l'Ohio, 1990.
Rumelhart De. and Hinton Wiliam, « Learning Internal Representations by Error Propagation », Explorations in the microstructure of cognition, Vol 1, pp. 318-362, Cambridge MIT Press, 1986.
Shephard GM, Synaptic organisation of the brain network, Oxford University Press, 1990.