Analyse syntaxique de surface

LAnalyse syntaxique de surface (ou « Analyse syntaxique superficielle » ; anglais shallow parsing ou light parsing) est une analyse syntaxique qui identifie les constituants d’une phrase (groupes nominaux, verbes…) sans spécifier leurs structures internes, ni leurs fonctions dans la phrase. Typiquement, l’analyse syntaxique de surface ne produit pas un arbre syntaxique complet. C’est une technique très utilisée en traitement automatique du langage naturel, ainsi que dans le test de lisibilité de Flesch. Notes et références Voir aussi

Analyse syntaxique

Catégorie:Syntaxe Catégorie:Intelligence artificielle

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Publications associées (3)

Publications associées

Chargement

Personnes associées

Chargement

Unités associées

Chargement

Concepts associés (1)

Concepts associés

Chargement

Cours associés (2)

Cours associés

Chargement

Séances de cours associées (3)

Séances de cours associées

Chargement

Traitement automatique du langage naturel

Le traitement automatique du langage naturel (TALN), en anglais natural language processing ou NLP, est un domaine multidisciplinaire impliquant la li

Automatic Segmentation of Nuclei in Histopathology Images using Encoding-Decoding Convolutional Neural Networks

Beril Besbinar, Pascal Frossard, Deniz Sayin Mercadier

Accurate and fast segmentation of nuclei in histopathological images plays a crucial role in cancer research for detection and grading, as well as personal treatment. Despite the important efforts, current algorithms are still suboptimal in terms of speed, adaptivity and generalizability. Popular Deep Convolutional Neural Networks (DCNNs) have recently been utilized for nuclei segmentation, outperforming traditional approaches that exploit color and texture features in combination with shallow classifiers or segmentation algorithms. However, DCNNs need large annotated datasets that require extensive amount of time and expert knowledge. In addition, segmentation results obtained by either traditional or DCNN approaches often require a post-processing step to separate cluttered nuclei. In this paper, we propose a computationally efficient nuclei segmentation framework based on DCNNs exhibiting an encoding-decoding structure. We use a partially-annotated dataset and develop an effective training solution. We also use a weighted background model for network to give more importance to borders of nuclei to overcome the problem of clutters. The abolition of any pre-processing or post-processing step without any compromise on the performance leads to a fast and parameter-free system, which presents important advantages with respect to state-of-the-art.

IEEE2019

Chargement

Lightweight Cross-Lingual Sentence Representation Learning

Prakhar Gupta, Martin Jaggi, Zhuoyuan Mao

Large-scale models for learning fixed-dimensional cross-lingual sentence representations like LASER (Artetxe and Schwenk, 2019b) lead to significant improvement in performance on downstream tasks. However, further increases and modifications based on such large-scale models are usually impractical due to memory limitations. In this work, we introduce a lightweight dual-transformer architecture with just 2 layers for generating memory-efficient cross-lingual sentence representations. We explore different training tasks and observe that current cross-lingual training tasks leave a lot to be desired for this shallow architecture. To ameliorate this, we propose a novel cross-lingual language model, which combines the existing single-word masked language model with the newly proposed cross-lingual token-level reconstruction task. We further augment the training task by the introduction of two computationally-lite sentence-level contrastive learning tasks to enhance the alignment of cross-lingual sentence representation space, which compensates for the learning bottleneck of the lightweight transformer for generative tasks. Our comparisons with competing models on cross-lingual sentence retrieval and multilingual document classification confirm the effectiveness of the newly proposed training tasks for a shallow model.

ASSOC COMPUTATIONAL LINGUISTICS-ACL2021

Chargement

Scala-Virtualized: Linguistic Reuse for Deep Embeddings

Nada Amin, Philipp Haller, Martin Odersky, Tiark Rompf

Scala-Virtualized extends the Scala language to better support hosting embedded DSLs. Scala is an expressive language that provides a flexible syntax, type-level computation using implicits, and other features that facilitate the development of em- bedded DSLs. However, many of these features work well only for shallow embeddings, i.e. DSLs which are implemented as plain libraries. Shallow embeddings automatically profit from features of the host language through linguistic reuse: any DSL expression is just as a regular Scala expression. But in many cases, directly executing DSL programs within the host language is not enough and deep embeddings are needed, which reify DSL programs into a data structure representation that can be analyzed, optimized, or further translated. For deep embeddings, linguistic reuse is no longer automatic. Scala-Virtualized defines many of the language’s built-in constructs as method calls, which enables DSLs to redefine the built-in semantics using familiar language mechanisms like overloading and overriding. This in turn enables an easier progression from shallow to deep embeddings, as core language constructs such as conditionals or pattern matching can be redefined to build a reified representation of the operation itself. While this facility brings shallow, syntactic, reuse to deep embeddings, we also present examples of what we call deep linguistic reuse: combining shallow and deep components in a single DSL in such a way that certain features are fully implemented in the shallow embedding part and do not need to be reified at the deep embedding level.

Springer Verlag2013

Chargement

EE-311: Fundamentals of machine learning

Ce cours présente une vue générale des techniques d'apprentissage automatique, passant en revue les algorithmes, le formalisme théorique et les protocoles expérimentaux.

CS-401: Applied data analysis

This course teaches the basic techniques, methodologies, and practical skills required to draw meaningful insights from a variety of data, with the help of the most acclaimed software tools in the data science world (pandas, scikit-learn, Spark, etc.)