Filters








141 Hits in 5.6 sec

An empirical study of Vietnamese noun phrase chunking with discriminative sequence models

Le Minh Nguyen, Huong Thao Nguyen, Phuong Thai Nguyen, Tu Bao Ho, Akira Shimazu
2009 Proceedings of the 7th Workshop on Asian Language Resources - ALR7   unpublished
This paper presents an empirical work for Vietnamese NP chunking task. We show how to build an annotation corpus of NP chunking and how discriminative sequence models are trained using the corpus.  ...  Experiment results using 5 fold cross validation test show that discriminative sequence learning are well suitable for Vietnamese chunking.  ...  This paper is supported by JAIST Grant for Research Associates and a part from a national project named Building Basic Resources and Tools for Vietnamese Language and Speech Processing, KC01.01/06-10.  ... 
doi:10.3115/1690299.1690301 fatcat:yhww6pgihbgixecxvhjibpnmvy

Getting Past the Language Gap: Innovations in Machine Translation [chapter]

Rodolfo Delmonte
2012 Mobile Speech and Advanced Natural Language Solutions  
Trust Rating % A generic typical pipeline for an SMT system using MOSES Discriminative models have been introduced so that translations are ranked and learned automatically by the use of features.  ...  Phrase-based SMT works as follows: source input is segmented in phrases (any sequence of words); each source phrase is automatically aligned to a target phrase on the basis of word alignment; and, eventually  ...  The experiment shows that the discriminative soft syntactic constraint model achieves better result over the heuristic model of M&R 2008; then, it empirically proves that the more accurate classifier can  ... 
doi:10.1007/978-1-4614-6018-3_6 fatcat:2njkc6meabhaxosl4wircumfjm

REFLECTIONS ON FREQUENCY EFFECTS IN LANGUAGE PROCESSING

Nick C. Ellis
2002 Studies in Second Language Acquisition  
patterns, and more extensive empirical studies of frequency and other factors which may affect noticing.  ...  For the case of second language phonology, there are the studies training Japanese L1 speakers on /r/ϳ/l/ discrimination.  ... 
doi:10.1017/s0272263102002140 fatcat:7zlauh2akbdplag55e6upwk25i

K(case)Ps: 'configurationality' and 'structural simplification' (pp. 85-97); What is 'lateral' grammaticalization? (pp. 98-110)

Keith Tse
2017 Figshare  
Neither of them analyses the grammaticalization of KPs (case-markers) and so this is the theme of this paper.  ...  Roberts and Roussou (2003) analyse grammaticalization within Minimalism, and Ledgeway (2011a, 2011b) deals with grammaticalization in Latin/Romance, also within Minimalism.  ...  There is a dialectical relationship between the language of policy documents and the perception of immigrants by the majority and their self-perception.  ... 
doi:10.6084/m9.figshare.5240788.v3 fatcat:vhl2m3vwqbavnmtb5dczjxcloi

LEPOR: An Augmented Machine Translation Evaluation Metric [article]

Aaron Li-Feng Han
2017 arXiv   pre-print
Firstly, we design augmented factors to yield highly accurate evaluation.Secondly, we design a tunable evaluation model where weighting of factors can be optimised according to the characteristics of languages  ...  One important issue in MT is that how to evaluate the MT system reasonably and tell us whether the translation system makes an improvement or not.  ...  Other similar works using the phrase similarity include the that uses noun phrase and verb phrase from chunking and (Echizen-ya and Araki, 2010) that only uses the noun phrase chunking in automatic evaluation  ... 
arXiv:1703.08748v1 fatcat:ifz4vscjhnclnartp7p5hzfx5u

POLYGLOT-NER: Massive Multilingual Named Entity Recognition [article]

Rami Al-Rfou, Vivek Kulkarni, Bryan Perozzi, Steven Skiena
2014 arXiv   pre-print
In this paper, we demonstrate how to build massive multilingual annotators with minimal human expertise and intervention.  ...  The increasing diversity of languages used on the web introduces a new level of complexity to Information Retrieval (IR) systems.  ...  Discriminative Learning We model NER as a word level classification problem.  ... 
arXiv:1410.3791v1 fatcat:kqkxgidkgzf2lp4iuh4twuicnm

Modeling Language Variation and Universals: A Survey on Typological Linguistics for Natural Language Processing

Edoardo Maria Ponti, Helen O'Horan, Yevgeni Berzak, Ivan Vulić, Roi Reichart, Thierry Poibeau, Ekaterina Shutova, Anna Korhonen
2019 Computational Linguistics  
We present an extensive literature survey on the use of typological information in the development of NLP techniques.  ...  We show that this is due to both intrinsic limitations of databases (in terms of coverage and feature granularity) and under-employment of the typological features included in them.  ...  For instance, the Ono language (Trans-New Guinea) expresses it with a lexical, monosyndetic, pre-nominal marker so in noun phrases.  ... 
doi:10.1162/coli_a_00357 fatcat:cfekqbpmwzegdf6j6atez2rsbe

The CoreGram project: theoretical linguistics, theory development and verification

Stefan Müller
2015 Journal of Language Modelling  
with other multilingual grammar engineering projects. overview and motivation The goal of the CoreGram project is to contribute to a better understanding of the constraints for specific human languages  ...  The paper discusses some of the highlights in the implemented grammars, gives a brief overview of central theoretical concepts and their implementation in the TRALE system, and compares the Core-Gram project  ...  Rather, I agree with Bender and Flickinger (1999) and Bender (2008, p. 20-21) that studying phenomena that are traditionally assigned to the periphery may discriminate between possible analyses of  ... 
doi:10.15398/jlm.v3i1.91 fatcat:syqpalp7fvdwbionla3szotggm

The Discriminative Lexicon: A Unified Computational Model for the Lexicon and Lexical Processing in Comprehension and Production Grounded Not in (De)Composition but in Linear Discriminative Learning

R. Harald Baayen, Yu-Ying Chuang, Elnaz Shafaei-Bajestan, James P. Blevins
2019 Complexity  
The discriminative lexicon also incorporates the insight from machine learning that end-to-end modeling is much more effective than working with a cascade of models targeting individual subtasks.  ...  The discriminative lexicon is introduced as a mathematical and computational model of the mental lexicon.  ...  This research was supported by an ERC advanced Grant (no. 742545) to the first author.  ... 
doi:10.1155/2019/4895891 fatcat:t4hlevfsfre4rc6bbllpcyja5a

Shallow features as indicators of English–German contrasts in lexical cohesion

Kerstin Kunz, Ekaterina Lapshinova-Koltunski, José Manuel Martínez Martínez, Katrin Menzel, Erich Steiner
2017 Languages in Contrast: International Journal for Contrastive Linguistics  
After an overview of research aims and motivations we formulate hypotheses on distributions of shallow features as indicators of lexical cohesion across languages and modes and with respect to register  ...  Descriptive analyses of the corpus are then presented and statistically validated with the help of univariate and multivariate analyses.  ...  This distinction is out of the scope of the syntactic phase as it is a target of the overall process of frame annotation.  ... 
doi:10.1075/lic.16005.kun fatcat:p7pa2crm6rgkrfmsdf7m5265ea

A Survey on Deep Learning for Named Entity Recognition [article]

Jing Li, Aixin Sun, Jianglei Han, Chenliang Li
2020 arXiv   pre-print
Early NER systems got a huge success in achieving good performance with the cost of human engineering in designing domain-specific features and rules.  ...  Finally, we present readers with the challenges faced by NER systems and outline future directions in this area.  ...  Instead of supervision, their model resorts to terminologies, corpus statistics (e.g., inverse document frequency and context vectors) and shallow syntactic knowledge (e.g., noun phrase chunking).  ... 
arXiv:1812.09449v3 fatcat:36tnstbyo5h4xizjpqn4cevgui

I * English Language

V. Haser, A. Auer, B. Botma, M. Elenbaas, W. Van Der Wurff, B. Gyuris, K. Allan, E. Vobornik, L. Anderwald, A. Schroder, C. Lozano, C. Hulmbauer (+5 others)
2013 Year's Work in English Studies  
study of L1 Vietnamese-L2 English learners.  ...  This volume contains theoretical, empirical and pedagogical papers. Of interest are the studies in section 1, which deal with L2 English phonological interlanguage.  ...  study of L1 Vietnamese-L2 English learners.  ... 
doi:10.1093/ywes/mat006 fatcat:peknqcgirzdmnlbpo3ethx6dyy

Arabic Named Entity Recognition: A Feature-Driven Study

Yassine Benajiba, Mona Diab, Paolo Rosso
2009 IEEE Transactions on Audio, Speech, and Language Processing  
Part-Of-Speech (POS) tags and Base Phrase Chunks (BPC): To derive Part of speech tags (POS) and base phrase chunks (BPC) we employ the AMIRA-1.0 system 15 described in I CHARACTERISTICS OF NLE-CORPUS  ...  Named Entity Recognition with Character-Level Models. Proc. of CoNLL-2003. [Kudo and Matsumato2000] Taku Kudo and Yuji Matsumato. 2000. Chunking with Support Vector Machine.  ... 
doi:10.1109/tasl.2009.2019927 fatcat:35z6fxuvubfnte5x4avyrs67xe

Understanding Disorder Within Variation: Production of English Grammatical Forms by English Language Learners

Lisa M. Bedore, Elizabeth D. Peña, Jissel B. Anaya, Ricardo Nieto, Mirza J. Lugo-Neris, Alisa Baron
2018 Language, Speech & Hearing Services in Schools  
Purpose: This study examines English performance on a set of 11 grammatical forms in Spanish-English bilingual, school-age children in order to understand how item difficulty of grammatical constructions  ...  In addition, the authors thank Stephanie McMillen for her assistance with data analysis.  ...  for Bilingual (Spanish-English) Children (NICHD R21HD53223, PI: Peña), and Cross-Language Outcomes of Typical and Atypical Development in Bilinguals (NIDCD 1 Bedore et al.: Production of English Grammatical  ... 
doi:10.1044/2017_lshss-17-0027 pmid:29621806 pmcid:PMC6105132 fatcat:dz6635som5c3tomlygbyewf7h4

Topic Segmentation and Labeling in Asynchronous Conversations

S. Joty, G. Carenini, R. T. Ng
2013 The Journal of Artificial Intelligence Research  
Empirical evaluation shows that the segmentation and the labeling performed by our best models beat the state-of-the-art, and are highly correlated with human annotations.  ...  Our approach extends state-of-the-art methods by considering a fine-grained structure of an asynchronous conversation, along with other conversational features by applying recent graph-based methods for  ...  Acknowledgments This work was conducted at the University of British Columbia.  ... 
doi:10.1613/jair.3940 fatcat:s264hyayknfpjiaqorgo67f32e
« Previous Showing results 1 — 15 out of 141 results