Natividad Prieto

Tagging and chunking with bigrams

In this paper we present an integrated system for tagging and chunking texts from a certain language. The approach is based on stochastic finite-state models that are learnt automatically. This includes biagram models or finite-state automata learnt using grammatical inference techniques. As the models involved in our system are learnt automatically, this is a very flexible and portable system.In order to show the viability of our approach we present results for tagging and chunking using bigram models on the Wall Street Journal corpus. We have achieved an accuracy rate for tagging of 96.8%, and a precision rate for NP chunks of 94.6% with a recall rate of 93.6%.

Learning language models through the ECGI method

Abstract A new approach to adaptive Semantic-Language modelling has recently been proposed which allows automatic learning of all the acoustic and syntactic-semantic models that are required for a given Continuous Speech Recognition task. The proposed approach is based on the so called “Error Correcting Grammatical Inference” algorithm which supplies homogeneous finite state structural models both at the acoustic and at the syntactic-semantic levels. Recognition or Understanding is seen as a Formal Transduction procedure that exploits the set of acoustic and linguistic constraints that have been captured in the learned models to directly input raw acoustic signals and output the semantic messages that are conveyed by these signals. In this paper, the proposed approach is reviewed and new improvements are presented. Also, preliminary results with a large semantic-space continuous speech task (Spanish numbers in the one-million range) are presented showing the currently achieved capabilities of this approach.

Improving chunking by means of lexical-contextual information in statistical language models

In this work, we present a stochastic approach to shallow parsing. Most of the current approaches to shallow parsing have a common characteristic: they take the sequence of lexical tags proposed by a POS tagger as input for the chunking process. Our system produces tagging and chunking in a single process using an Integrated Language Model (ILM) formalized as Markov Models. This model integrates several knowledge sources: lexical probabilities, a contextual Language Model (LM) for every chunk, and a contextual LM for the sentences. We have extended the ILM by adding lexical information to the contextual LMs. We have applied this approach to the CoNLL-2000 shared task improving the performance of the chunker.

An Integrated Statistical Model for Tagging and Chunking Unrestricted Text

In this paper, we present a corpus-based approach for tagging and chunking. The formalism used is based on stochastic finite-state automata. Therefore, it can include n-grams models or any stochastic finite-state automata learnt using grammatical inference techniques. As the models involved in our system are learnt automatically, it allows for a very flexible and portable system for different languages and chunk definitions. In order to show the viability of our approach, we present results for tagging and chunking using different combinations of bigrams and other more complex automata learnt by means of the Error Correcting Grammatical Inference (ECGI) algorithm. The experimentation was carried out on the Wall Street Journal corpus for English and on the Lexesp corpus for Spanish.

Application of the error correcting grammatical inference method (ECGI) to multi-speaker isolated word recognition

It is well known that speech signals constitute highly structured objects which are composed of different kinds of subobjects such as words, phonemes, etc. This fact has motivated several researchers to propose different models which more or less explicitly assume the structural nature of speech. Notable examples of these models are Markov models /Bak 75/, /Jel 76/; the famous Harpy /Low 76/; Scriber and Lafs /Kla 80/; and many others works in which the convenience of some structural model of the speech objects considered is explicitly claimed /Gup 82/, /Lev 83/, /Cra 84/, /Sca 85/, /Kam 85/, /Sau 85/, /Rab 85/, /Kop 85/, /Sch 85/, /Der 86/, /Tan 86/.

Support planning to the implementation of School of Engineering in Computer Science - An oriented approach to teacher training

Main objective of Action Plan for European Convergence (PACE) that Universitat Politecnica de Valencia (UPV) has been developing since academic year 2005-2006 has been to facilitate the process of adaptation of new courses of bachelors degree to the European Higher Education Area (EEES). To carry out this ambitious project, centers have identified a set of coordinated actions by teams of teachers interested in teaching innovation. One of the activities promoted by School of Engineering in Computer Science (ETSINF) of UPV is a training plan for teachers. The main objective is to provide teachers with tools that facilitate the design of new courses of Bachelors Degree in Computer Engineering under the current regulations, with particular emphasis on two aspects: the use of active methodologies and the evaluation criteria. This article describes the training activities that have taken place in the ETSINF during academic year 2009-2010 and the results of opinion poll for teachers about the appropriateness of these actions and their impact on the process to adapt their courses.