Search Results

You are looking at 1 - 10 of 11 items

  • Author: Mark Davies x
Clear All Modify Search

Abstract Within the past decade several large, freely-available online corpora of Spanish and Portuguese have become available. With these new corpora, researchers of Spanish and Portuguese can now carry out the same type of corpus-based research that has been done for other languages (such as English) for years. This includes advanced research on morphological and syntactic variation (thanks to full functionality with substring searches, part of speech tagging, and lemmatization), semantics and pragmatics (via collocates, synonyms, customized word lists, and word comparisons), and historical changes and synchronic register variation (via architectures and interfaces that allow easy comparisons of frequency in different sections of the corpus).


This paper provides many concrete examples of how English grammar varies in important ways, as a function of differences between genres, as a function of language change, and as a function of differences between dialects. We also show – in great detail – how several recent corpora – such as COCA (2008), COHA (2011), GloWbE (2013), and the BYU interface to the Google Book n-grams (2012) – allow us to accurately examine this full range of variation, in ways that are not available with smaller corpora. As a result, these new corpora allow us to provide a much more reliable and insightful view into English syntax than was possible even four or five years ago.


The syntactic and semantic properties of the into-causative construction, involving the pattern ‘V NP into V-ing’, has been studied in some detail, but its historical development has received less attention. This paper looks at the historical shifts in the construction with much more representative and robust corpora than have been used for previous studies. Based on nearly 11,440 tokens from more than 950 million words of text, the paper considers several important changes in the lexical, syntactic, and semantic properties of the construction. The study also establishes the value of large, diverse historical corpora and their role in researching syntactic and semantic change.

Antimicrobial Susceptibility of Bacteria Isolated from Newborns with Suspected or Confirmed Necrotising Enterocolitis

Aim. This audit aimed to identify which bacteria were associated with necrotising enterocolitis (NEC) and determine their antibiotic sensitivities.

Methods. A retrospective audit of all infants with a diagnosis of suspected NEC or confirmed NEC and a positive culture (blood/faeces/operative specimen or vascular access device) between January 2000 and September 2007 was performed.

Results. Ninety nine infants had a diagnosis of suspected (45) or confirmed NEC (54). Seventeen patients had suspected (5) or confirmed (12) NEC and a positive culture result. 12 babies had positive blood cultures associated with their NEC. Only 4 of the 12 cases of NEC with a positive blood culture received adequate first line cover for their subsequently identified infecting organism.

Conclusions. Due to the limitations of this study we are unable to make general recommendations on the first line antibiotic choice for babies with suspected or confirmed NEC. Our current regime of Ampicillin, Gentamicin and Metronidazole failed to adequately treat 8 of the 12 organisms subsequently isolated in blood cultures. Only the combination of Vancomycin and Meropenem would have adequately treated all the bacteria identified. The concern with this approach is the possible emergence of multi drug resistant bacteria.


Spatial prediction is an important problem in many scientific disciplines. Super Learner is an ensemble prediction approach related to stacked generalization that uses cross-validation to search for the optimal predictor amongst all convex combinations of a heterogeneous candidate set. It has been applied to non-spatial data, where theoretical results demonstrate it will perform asymptotically at least as well as the best candidate under consideration. We review these optimality properties and discuss the assumptions required in order for them to hold for spatial prediction problems. We present results of a simulation study confirming Super Learner works well in practice under a variety of sample sizes, sampling designs, and data-generating functions. We also apply Super Learner to a real world dataset.


Cold water immersion is thought to reduce the inflammatory response to injury. Using cultured mononuclear cells and human subjects in a mixed martial arts (MMA) contest, we examined the effect of cryotherapy on 7,8-dihydroneopterin and neopterin generation. Urine was collected from 10 elite male mixed martial artists before, immediately post and 1, 2, 24 and 48 h following a contest. Myoglobin was analysed by reverse-phase high performance liquid chromatography, and urinary neopterin and total neopterin (neopterin+7,8-dihydroneopterin) were measured by strong cation exchange high-performance liquid chromatography. Cold water immersion and passive recovery were compared using changes in these markers, while cryotherapy tested total neopterin production in γ-interferon and phorbol myristate acetate (PMA)-stimulated blood-derived mononuclear cells (monocytes/T cells). Myoglobin significantly increased (p<0.05) at 1 h post-contest, neopterin significantly increased at 1 and 24 h (p<0.05), total neopterin significantly increased (p<0.05) at 1 h post for the passive group only, and significant individual variation was observed for all markers (p<0.01). Cold water immersion attenuated total neopterin production (p<0.05), while cryotherapy significantly reduced total neopterin production in PMA-stimulated mononuclear cells (p<0.01). Cryotherapy attenuates the post-exercise inflammatory response following an MMA contest. The evidence also suggests that the mechanisms responsible for this may be related to direct immune cell suppression.