Motivated by phrase-based translation research, we present a transliteration system where characters are grouped into substrings to be mapped atomically into the target language. We show how this substring representation can be incorporated into a Conditional Random Field model that uses local context and phonemic information.

RÉSUMÉ. Le travail présenté dans cet article vise à extraire automatiquement certaines caractéristiques de produits à partir de descriptions textuelles fournies par un site marchand. La constitution d’un corpus de référence annoté révèle certains problèmes, provenant à la fois des textes et des particularités de la tâche. Pour l’aborder, nous avons testé deux approches : une méthode d’extractio...

Intonation phrase length distribution is important information for prosodic break prediction. However, existing CRF frameworks cannot make full use of it. An expanded CRF is proposed in this paper to tackle this problem. Its lattice carries the location of previous intonation phrase (L3) break, and consequently makes it possible to support various dynamic features, such as the number of syllabl...

Article history: Received 18 May 2007 Received in revised form 1 October 2008 Accepted 24 May 2009

Area sources are a key contributor to overall greenhouse gas emissions but present a particular challenge to emission measurement techniques due to the heterogeneous nature of the sources. A new Controlled Release Facility (CRF) has been developed that is able to recreate in the field both the distribution and rate of emissions seen in actual industrial applications. The results of a series of ...

The Center for Disease Control and Prevention’s Active Bacterial Core Surveillance (CDC ABCs) Program is a collaborative effort betweeen the CDC, state health departments, laboratories, and universities to track invasive bacterial pathogens of particular importance to public health [1]. The year-end surveillance reports produced by this program help to shape public policy and coordinate respons...

We present a model for automatically predicting information status labels for German referring expressions. We train a CRF on manually annotated phrases, and predict a fine-grained set of labels. We achieve an accuracy score of 69.56% on our most detailed label set, 76.62% when gold standard coreference is available.

We present the Crotal system, used in the CoNLL09 Shared Task. It is based on XCRF, a highly configurable CRF library which can take into account hierarchical relations. This system had never been used in such a context thus the performance is average, but we are confident that there is room for progression.

viii Εκτεταμένη Περίληψη x

We introduce fast millimeter-wave base station (BS) and its antenna sector selection for user equipment based on its location. Using a conditional random field inference model with specially designed parameters, which are robust to change of environment, InferBeam allows the use of measurement samples on best beam selection at a small number of locations to infer the rest dynamically. Compared ...