The reflexive (among other values) french morpheme 'se' is studied here from an onomasiologic per... more The reflexive (among other values) french morpheme 'se' is studied here from an onomasiologic perspective. It is shown that its values can be derived from an operationalist / functionalist framework.
Encyclopédie Philosophique Universelle, Volume Notions, tome 2, 1990
The fundamental cognitive and linguistic operations characteristic of the noun (substantive, adje... more The fundamental cognitive and linguistic operations characteristic of the noun (substantive, adjective) are described here.
22 congrès Linguistique et Philologie Romane, tome 6, Max Niemeyer, 2000
Starting with the question of the variable adjective order in French this paper proposes a theori... more Starting with the question of the variable adjective order in French this paper proposes a theorized description of categories and functions in nominal syntagm, in a structuralist, functionalist, operationalist and corpus-based framework.
Études sur l'ordre des mots, collection ERA 642, 1988
This paper studies the variability of position in modern French adjectives. Its purpose is to giv... more This paper studies the variability of position in modern French adjectives. Its purpose is to give a functional, enunciative and formalistic analysis, showing that the function of adjectives with respect to the substantive determines their position relative to it.
It first shows the inadequacies and unsufficiencies of the main theories and remarks on the subject, then gives a classificatory description of adjectives according to their position and semantics. Built upon this classification and a study of text occurrences, the conclusion determines the two possible functions of adjectives : on the left side, situation-oriented adjectives, on the right side, specifying adjectives.
Encyclopédie Philosophie Universelle, Volume Notions, tome 2, 1990
The ancient problematic covered by the word predication as used by logicists, grammarians and lin... more The ancient problematic covered by the word predication as used by logicists, grammarians and linguists is described with its consequences on the development of modern concepts.
Encyclopédie Philosophique Universelle, Volume Notions, tome 2, 1990
Encyclopedic presentation of the enunciation problematics, with a tentative systematization of it... more Encyclopedic presentation of the enunciation problematics, with a tentative systematization of its fundamental concepts.
This paper presents the elements and principles of a non treelike dynamic view of syntax and the ... more This paper presents the elements and principles of a non treelike dynamic view of syntax and the interaction between grammatical and lexical knowledge.
Formalisation des relations prédicatives, Collection ERA 642, 1987
This paper presents very concretely on French examples the choices to be made when devising a me... more This paper presents very concretely on French examples the choices to be made when devising a mecanism for analysing and understanding sentences, as applied to French in particular. The methodology presented is functionalist, operationalist and resolutely privilegies an onomasiologic approach.
The reflexive (among other values) french morpheme 'se' is studied here from an onomasiologic per... more The reflexive (among other values) french morpheme 'se' is studied here from an onomasiologic perspective. It is shown that its values can be derived from an operationalist / functionalist framework.
Encyclopédie Philosophique Universelle, Volume Notions, tome 2, 1990
The fundamental cognitive and linguistic operations characteristic of the noun (substantive, adje... more The fundamental cognitive and linguistic operations characteristic of the noun (substantive, adjective) are described here.
22 congrès Linguistique et Philologie Romane, tome 6, Max Niemeyer, 2000
Starting with the question of the variable adjective order in French this paper proposes a theori... more Starting with the question of the variable adjective order in French this paper proposes a theorized description of categories and functions in nominal syntagm, in a structuralist, functionalist, operationalist and corpus-based framework.
Études sur l'ordre des mots, collection ERA 642, 1988
This paper studies the variability of position in modern French adjectives. Its purpose is to giv... more This paper studies the variability of position in modern French adjectives. Its purpose is to give a functional, enunciative and formalistic analysis, showing that the function of adjectives with respect to the substantive determines their position relative to it.
It first shows the inadequacies and unsufficiencies of the main theories and remarks on the subject, then gives a classificatory description of adjectives according to their position and semantics. Built upon this classification and a study of text occurrences, the conclusion determines the two possible functions of adjectives : on the left side, situation-oriented adjectives, on the right side, specifying adjectives.
Encyclopédie Philosophie Universelle, Volume Notions, tome 2, 1990
The ancient problematic covered by the word predication as used by logicists, grammarians and lin... more The ancient problematic covered by the word predication as used by logicists, grammarians and linguists is described with its consequences on the development of modern concepts.
Encyclopédie Philosophique Universelle, Volume Notions, tome 2, 1990
Encyclopedic presentation of the enunciation problematics, with a tentative systematization of it... more Encyclopedic presentation of the enunciation problematics, with a tentative systematization of its fundamental concepts.
This paper presents the elements and principles of a non treelike dynamic view of syntax and the ... more This paper presents the elements and principles of a non treelike dynamic view of syntax and the interaction between grammatical and lexical knowledge.
Formalisation des relations prédicatives, Collection ERA 642, 1987
This paper presents very concretely on French examples the choices to be made when devising a me... more This paper presents very concretely on French examples the choices to be made when devising a mecanism for analysing and understanding sentences, as applied to French in particular. The methodology presented is functionalist, operationalist and resolutely privilegies an onomasiologic approach.
More than 35 years ago, Z. Harris introduced an algorithm for string analysis remarkable by its s... more More than 35 years ago, Z. Harris introduced an algorithm for string analysis remarkable by its simplicity et its lack of need for external knowledge: recurrent stochastic dependency process. Its principle is that as our knowledge of a string progresses, the predictability of its unknown part gets higher. The aim of this paper is to show the feasibility of applying this algorithm to extracting syntactic strings in a French raw corpus. In order to do this, we propose an implementation which corrects some of the drawbacks of the original algorithm.
This paper reassesses the relationship between linguistics and artificial intelligence, and revis... more This paper reassesses the relationship between linguistics and artificial intelligence, and revisits its recent history, based on the following hypotheses: (1) artificial intelligence is a social science; (2) in the no man's land between artificial intelligence and formal linguistics models appeared without a clear status playing a buffering role; (3) For linguistics assuming this heritage could lay ground for a redefinition of the paradigm of theoretical and experimental linguistics and of the opposition between natural and social sciences.
Stratégies théoriques, Collection Histoire et Théories Linguistiques 10/1gie Langage, 1988
This paper presents a view on the relationship between linguistics and artificial intelligence. C... more This paper presents a view on the relationship between linguistics and artificial intelligence. Compare with my more recent paper on the same subject, La linguistique et l'intelligence artificielle.
Linx 17, Sur la transitivité dans les langues, Annie Montaut[ed sc], 1991
This paper presents a conception of the grammaticalization of transitivity phenomena that has dev... more This paper presents a conception of the grammaticalization of transitivity phenomena that has developed from the end of the 19th century until the 1960's (Breal, Meillet, Blinkenberg, Sapir, Kurylowicz). This conception has never been formalized as a theory, but nevertheless has constituted the background for numerous construction evolution analysis of linguists of this period.
« Verner’s may be the single most influential publication in linguistics. It is so lucid that it ... more « Verner’s may be the single most influential publication in linguistics. It is so lucid that it scarcely needs comment. » said Lehmann. Here is a reader's guide, following Verner's methodology, tying it with the indo-european reconstructions, and explaining the (nowadays) not so much known concepts used by Carl Verner.
Cahiers du GLECS (Groupe de Linguistique Chamito-sémitique), 2000
This paper is about the noun *qirb (lap) in afro-asiatic languages. After discussing the forms in... more This paper is about the noun *qirb (lap) in afro-asiatic languages. After discussing the forms included in the original classification of Marcel Cohen, a reconstruction of the evolution of the word is proposed, including an explanation for the arabic anomalous form qalb.
It then proceeds to the reconstruction of the meanings of the word and its derivative, founded on semantic parallelisms, and finally proposes some paths for linking these meanings to the qrb verb.
Banks are compelled by financial regulatory authorities to demonstrate whole-hearted commitment t... more Banks are compelled by financial regulatory authorities to demonstrate whole-hearted commitment to finding ways of preventing suspicious activities. Can AI help monitor user behavior in order to detect fraudulent activity such as identity theft? In this paper, we propose a Machine Learning (ML) based fraud detection framework to capture fraudulent behavior patterns and we experiment on a real-world dataset of a major European bank. We gathered recent state-of-the-art techniques for identifying banking fraud using ML algorithms and tested them on an abnormal behavior detection use case.
2021 IEEE 33rd International Conference on Tools with Artificial Intelligence (ICTAI), Nov 1, 2021
This paper studies the problem of applying artificial intelligence (AI) methods to profiling fina... more This paper studies the problem of applying artificial intelligence (AI) methods to profiling financial crime behavior, more specifically money laundering in a context of environmental trafficking. The main objective is to provide a tool based on Neural Networks (NNs) for identification and analysis of suspicious transactions. This study is part of a collaborative research project with HSBC and we paid particular attention to real-world financial industry constraints, like real-time fast processing and explainability.
2021 IEEE International Conference on Big Data (Big Data), Dec 15, 2021
This paper studies the problem of applying artificial intelligence (AI) methods to the detection ... more This paper studies the problem of applying artificial intelligence (AI) methods to the detection of financial Suspicious Activity Reports (SARs). Financial institutions use semi-automated systems based on predetermined rules to identify suspicious activities, but most of the generated alerts are false positives and do not lead to a reporting to the authorities. From the initial investigation based on a predefined triggering scenario, to the decision of a case worth reporting to the authorities, critical parts of the process are carried out by human analysts. Part of a collaborative project with HSBC, the main objective of this research is to automate the identification of truly suspicious transactions by reducing the need for human intervention, as well as limiting the number of false alerts, using Neural Networks (NNs). We implement a prediction framework based on a combination of two unsupervised NNs used in an online learning mode and we compare the performances of our solution against other Machine Learning (ML) models.
Les réseaux de neurones profonds dominent aujourd’hui dans des domaines comme la reconnaissance d... more Les réseaux de neurones profonds dominent aujourd’hui dans des domaines comme la reconnaissance d’image ou de la parole et le traitement du langage naturel (Krizhevsky et al., 2012; Goodfellow et al., 2016). Nous abordons ici la question de savoir s’il est possible de faire bénéficier des algorithmes plus simples, moins gourmands en mémoire, en puissance de calcul, en temps de paramétrage et en quantité de données d’apprentissage, des avantages de l’architecture profonde. Nous proposons dans cet article 1 une extension profonde des forêts aléatoires, appelée DET.
Notre etude presente la conception et la realisation d'un systeme d'analyse syntaxique de... more Notre etude presente la conception et la realisation d'un systeme d'analyse syntaxique de surface dans un grand corpus non annote en francais. Le systeme START (System of Textual Analysis Recognition and Tagging) effectue les tâches suivantes : determination des frontieres des constituants d'une phrase, extraction des groupes syntaxiques (groupes nominaux, verbaux, prepositionnels, etc. ), etiquetage partiel des mots qui composent des constituants detectes, desambiguisation des cas ambigus, creation d'un dictionnaire a partir des mots qui composent les constituants. Le systeme n'utilise aucun dictionnaire et il utilise un minimum de regles grammaticales et syntaxiques. Il est base sur des statistiques issues de l'analyse distributionnelle des mots grammaticaux. Le processus est novateur car il n'utilise pas de corpus etiquete et aucune connaissance prealable. Les resultats depassent 92% de reconnaissance pour les GV et des GN et 99,6% pour la desambiguisation, avec un taux d'erreur inferieur a 1%.
Deep neural networks have recently become popular because of their success in such domains as ima... more Deep neural networks have recently become popular because of their success in such domains as image and speech recognition, which has lead many to wonder whether other learners could benefit from deep, layered architectures. In this paper, we propose the Deep Cascade of Extra Trees (DCET) model. Representation learning in deep neural networks mostly relies on the layer-by-layer processing of raw features. Inspired by this, DCET uses a deep cascade of decision forests structure, where the cascade in each level receives the best feature information processed by the cascade of forests of its preceding level. Experiments show that its performance is quite robust regarding hyper-parameter settings; in most cases, even across different datasets from different domains, it is able to get excellent performance by using the same default setting.
2021 International Conference on Electrical, Computer and Energy Technologies (ICECET), 2021
A corpus is a large collection of spoken or written one or more language, it is collected from di... more A corpus is a large collection of spoken or written one or more language, it is collected from different resources, then structured, stored, and treated automatically by special algorithm to reach the desired goal, this is why it is used by researchers in different domain such as grammar, semantics, lexicography, natural Language Processing and other language studies. Therefore, building a corpus was still a challenge for many researchers in those fields for many years. Our study in this paper aims to build a new gold standard Arabic corpus due to the lack of successful trials in compiling Arabic corpora. The corpus produced by our team, is a text corpus, collected from a set of Arabic Newspaper articles morphologically analyzed from eight Arabic countries, and it contains more than 18 million words in total, covering six categories (Religion, Economy, Culture, Sports, Local and International News). It was encoded with UTF -8 encoding and marked with two mark-up languages: JSON and XML. In the hope that this corpus can be used as an accurate reference for segmentation and validation and learning in the syntax analysis mainly for the word segmentation and part of speech tagging.
Many syntactic parsers of Arabic language were developed during the last few years, but just few ... more Many syntactic parsers of Arabic language were developed during the last few years, but just few of them adopted the statistical approach and almost all of these parsers are based on supervised learning techniques. In this paper we present our efforts towards the development of a new statistical parser of Arabic language based on the unsupervised learning model.
Uploads
It first shows the inadequacies and unsufficiencies of the main theories and remarks on the subject, then gives a classificatory description of adjectives according to their position and semantics. Built upon this classification and a study of text occurrences, the conclusion determines the two possible functions of adjectives : on the left side, situation-oriented adjectives, on the right side, specifying adjectives.
It first shows the inadequacies and unsufficiencies of the main theories and remarks on the subject, then gives a classificatory description of adjectives according to their position and semantics. Built upon this classification and a study of text occurrences, the conclusion determines the two possible functions of adjectives : on the left side, situation-oriented adjectives, on the right side, specifying adjectives.
It then proceeds to the reconstruction of the meanings of the word and its derivative, founded on semantic parallelisms, and finally proposes some paths for linking these meanings to the qrb verb.