This paper considers the hypothesis that voting between multiple data representations can be more accurate than voting between multiple learning models. This hypothesis has been considered before (cf. [San00]) but the focus was on voting methods rather than the data representations. In this paper, we focus on choosing specific data representations combined with simple majority voting. On the co...

Resumen: Investigamos la posibilidad de obtener patrones semm anticos a gran es-cala para cualquier lengua usando solamente ann alisis superrcial y generalizaciones semm anticas bb asicas. Siendo este un experimento exploratorio ss olo hemos realizado una evaluacii on cualitativa. Hemos comparado varios patrones semm anticos de tra-duccii on de verbos equivalentes en distintas lenguas y dominio...

The SEMum model of Lekic and Romanowicz (2011) (SEMum v.1) was the first global VS model obtained using spectral-element forward modeling (SEM: Komatitsch and Vilotte, 1998), and exhibits impressive amplitudes of heterogeneity in the upper 200km of the mantle compared to previous global models. Among other measures to make SEM-based modeling tractable, SEMum v.1 was developed using an homogeniz...

Machine Learning of a chunker for French We describe in this paper how to automatically learn a chunker for French, from the French Tree Bank and CRFs (Conditional Random Fields). We did several experiments, either to recognize every possible kind of chunks, or to focus on simple nominal phrases only. We evaluate the obtained chunker on internal data (i.e. also extracted from the French Tree Ba...

Part-of-speech (POS) tagging and chunking have been used in tasks targeting learner English; however, to the best our knowledge, few studies have evaluated their performance and no studies have revealed the causes of POStagging/chunking errors in detail. Therefore, we investigate performance and analyze the causes of failure. We focus on spelling errors that occur frequently in learner English....

The length of a constituent (number of syllables in a word or number of words in a phrase), or rhythm, plays an important role in Chinese syntax. This paper systematically surveys the distribution of rhythm in constructions in Chinese from the statistical data acquired from a shallow tree bank. Based on our survey, we then used the rhythm feature in a practical shallow parsing task by using rhy...

Speech translation systems commonly couple automatic speech recognition (ASR) and machine translation (MT) components. Hereby the automatic segmentation of the ASR output for the subsequent MT is critical for the overall performance. In simultaneous translation systems, which require a continuous output with a low latency, chunking of the ASR output into translatable segments is even more criti...

A SNoW based learning approach to shallow parsing tasks is presented and studied experimentally. The approach learns to identify syntactic patterns by combining simple predictors to produce a coherent inference. Two instantiations of this approach are studied and experimental results for Noun-Phrases (NP) and Subject-Verb (SV) phrases that compare favorably with the best published results are p...

Blogs are highly rich in opinion making their automatic processing appealing to marketing companies, the media, costumer centres, etc. TREC ran a Blog track in 2006 with two tasks: opinion retrieval and an open task. This document reports the experiments conducted at The Robert Gordon University (RGU) where we used Statistical Language Models combined with shallow parsing techniques for the opi...

We apply a uniform Cramer-Rao (CR) bound [l] to study the bias-variance trade-offs in parameter estimation. The uniform CR bound is used to specify achievable and unachievable regions in the bias-variance trade-off plane. The applications considered in this paper are: 1) twodimensional single photon emission computed tomography (SPECT) system, and 2) one dimensional edge localization.