We describe a transfer method based on annotation projection to develop a dependency-based semantic role labeling system for languages for which no supervised linguistic information other than parallel data is available. Unlike previous work that presumes the availability of supervised features such as lemmas, part-of-speech tags, and dependency parse trees, we only make use of word and character features. Our deep model considers using character-based representations as well as unsupervised stem embeddings to alleviate the need for supervised features. Our experiments outperform a state-of-the-art method that uses supervised lexico-syntactic features on 6 out of 7 languages in the Universal Proposition Bank.

TiO2 nanofibres were prepared using a templating method with tetraisopropylorthotitanate (TiPT) as a precursor. The preparation comprises liquid phase deposition on cellulose fibres followed by thermal removal of the cellulose template. The obtained TiO2 fibrous substance consists of micron-size fibres with a microstructure of nanofibres. It was demonstrated that nanofibres are basically formed through the aggregation of TiO2 nanoparticles and nanorods into chain structures during the thermal treatment process. The measured surface area of the TiO2 fibres was about 250 m2 g−1. It was shown that the pore size distribution is multi-scale and a fractal morphology was demonstrated with two fractal regimes with dimensions of 1.78 and 2.97 for sizes below and above 7.5 nm, respectively. The crystalline phase of the TiO2 nanofibres, as well as the nanoparticles in the solution, could be controlled by the pH of the solution. A pH of 1.2 resulted in rutile phase, while a pH of 1.8 resulted in anatase phase.

Dialects and standard forms of a language typically share a set of cognates that could bear the same meaning in both varieties or only be shared homographs but serve as faux amis. Moreover, there are words that are used exclusively in the dialect or the standard variety. Both phenomena, faux amis and exclusive vocabulary, are considered out of vocabulary (OOV) phenomena. In this paper, we present this problem of OOV in the context of machine translation. We present a new approach for dialect to English Statistical Machine Translation (SMT) enhancement based on normalizing dialectal language into standard form to provide equivalents to address both aspects of the OOV problem posited by dialectal language use. We specifically focus on Arabic to English SMT. We use two publicly available dialect identification tools: AIDA and MADAMIRA, to identify and replace dialectal Arabic OOV words with their modern standard Arabic (MSA) equivalents. The results of evaluation on two blind test sets show that using AIDA to identify and replace MSA equivalents enhances translation results by 0.4% absolute BLEU (1.6% relative BLEU) and using MADAMIRA achieves 0.3% absolute BLEU (1.2% relative BLEU) enhancement over the baseline. We show our replacement scheme reaches a noticeable enhancement in SMT performance for faux amis words.

Background:Herpes Zoster (HZ) is reactivation of latent varicella-zoster virus that involves dermatomes. Aging and immunosupressed states are among the main risk factors. Some investigations showed that HZ is more common in diabetic patients than in normal population.Aim:To determine whether undiagnosed DM is more common in patients with HZ than in those without it.Materials and Methods:In this study 103 patients with HZ (cases) and 142 as control participated. They had no history of DM. Both groups were matched according to age, gender and family history of DM. Fasting plasma glucose was checked for all participants. DM was defined when the fasting plasma glucose was equal or more 126 mg/dl.Results:35.9% of patients with HZ and 19.7% of the control group had DM. There was significant association between HZ and undiagnosed DM (OR = 2.28, 95% CI: 1.28–4.06).Conclusion:Our findings indicate that the prevalence of undiagnosed DM is more common in HZ patients and supports the policy to investigate patients with HZ for the presence of undiagnosed DM.

We describe a method for developing broadcoverage semantic dependency parsers for languages for which no semantically annotated resource is available. We leverage a multitask learning framework coupled with annotation projection. We use syntactic parsing as the auxiliary task in our multitask setup. Our annotation projection experiments from English to Czech show that our multitask setup yields 3.1% (4.2%) improvement in labeled F1-score on in-domain (out-of-domain) test set compared to a single-task baseline.