Combining automatic acquisition of knowledge with machine learning approaches for multilingual temporal recognition and normalization

Saquete, Estela; Ferrández, Óscar; Ferrández, Sergio; Martínez-Barco, Patricio; Muñoz, Rafael

doi:10.1016/j.ins.2008.05.002

Cited by 10 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of tools, only a multilingual POSTAGGER, a multilingual signal recognizer and a tool that normalizes temporal expressions at a multilingual level are required. In fact, the T2T3 transducer implemented uses a multilingual POSTAGGER (MACO in Carmona et al 1998) and multilingual normalizer (TERSEO in Saquete et al 2008) in order to migrate easily to other languages. However, the adaptation of the extent is a more language-dependent task, and although the main relations between expressions appear in the different languages (specification, anchoring, range and conjunction), there are some specific syntactic or semantic situations that should be treated in detail, for example, the possessive constructions, which have different structures in other languages.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic transformation from TIDES to TimeML annotation

Saquete

Pustejovsky

2011

Lang Resources & Evaluation

View full text Add to dashboard Cite

Until recently, most systems performing temporal extraction and reasoning from text have focused on recognizing and normalizing temporal expressions alone, for which the TIDES annotation scheme has been adopted. Temporal awareness of a text, however, involves not only identifying the temporal expressions, but the events which these expressions anchor, as well as other events which must be ordered relative to them. Because of these broader concerns, TimeML has been developed as an annotation specification that encompasses not only temporal expressions, but all temporally relevant aspects of a text. The annotation schemes, however, are not interchangeable, resulting in incompatible corpora and accompanying extraction algorithms for each standard. In this paper, we describe an automatic migration process from the TIMEX2 tags of TIDES to the TIMEX3 tags of TimeML. This transformation procedure has been implemented and evaluated with two different corpora, obtaining 93.3 and 89.2% overall F-Measure respectively.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Some systems use rule-based approaches (Saquete et al 2006), whereas others are using machine learning approaches (Saquete et al 2008;TempEx 2008;Gerber et al 2002). Choosing the appropriate approach depends on the available resources and the requirements of the systems being developed.…”

Section: Introductionmentioning

confidence: 99%

Automatic transformation from TIDES to TimeML annotation

Saquete

Pustejovsky

2011

Lang Resources & Evaluation

View full text Add to dashboard Cite

show abstract

“…The term identification and NER are mainly performed using rule-based and machine learning approaches [45]. Fukuda [16] developed one of the earliest NER systems for proteins, but the rules had to be defined manually.…”

Section: Named Entity Recognition (Ner)mentioning

confidence: 99%

BioOntoVerb: A top level ontology based framework to populate biomedical ontologies from texts

Ruiz-Martínez

Valencia‐García

Martı́nez-Béjar

et al. 2012

Knowledge-Based Systems

View full text Add to dashboard Cite

“…ML commonly used in today's software is to study and apply the computer modeling of learning processes in their multiple manifestations, which can facilitate language learning and encourage the development of intelligent system [5][6]. The combined advantages of learning from examples and learning by observation benefit the development of ML which includes inductive decision trees and neural networks etc [7][8].…”

Section: Introductionmentioning

confidence: 99%

Machine Learning for Second Language Learning: Effects on Syntactic Processing in Garden-Path Sentences

2010

2010 International Conference on Computational Intelligence and Software Engineering

View full text Add to dashboard Cite

With the rapid development of educational technologies, machine learning (ML) based second language learning (SLL) attracts the attention of many scholars from computational linguistics. Garden path (GP) sentence is a special sentence structure in which processing breakdown and backtracking are involved in the machine decoding. Faced with GP sentence, learners have to make original misinterpretation linger even after reanalysis has occurred, and have to face the fact that initial interpretations of GP sentence are often incorrect and do not correspond to the final output. On the basis of computational skills of N-S flowchart, recursive transition network, and well-formed substring table, this paper highlights the effects on syntactic processing in garden-path sentences when context-free grammar (CFG) driven ML system is introduced. There are two main conclusions to be drawn from the above discussion. Firstly, CFG-driven ML system can help SLL learners better understand the GP sentences in second language even no rich lingual environment is provided. Secondly, the hybrid knowledge, e.g. grammatical, cognitive and semantic information, needs to be integrated into ML system to help SLL learners achieve the "good enough" results in decoding GP sentences.

show abstract

Combining automatic acquisition of knowledge with machine learning approaches for multilingual temporal recognition and normalization

Cited by 10 publications

References 36 publications

Automatic transformation from TIDES to TimeML annotation

Automatic transformation from TIDES to TimeML annotation

BioOntoVerb: A top level ontology based framework to populate biomedical ontologies from texts

Machine Learning for Second Language Learning: Effects on Syntactic Processing in Garden-Path Sentences

Contact Info

Product

Resources

About