Filters








37,178 Hits in 2.3 sec

Will worms really cure Crohn's disease?

G L Radford-Smith
2005 Gut  
Will worms really cure Crohn's disease?  ...  This is turn will be affected by (1) the genetic background of the individual, which will determine whether they generate strong regulatory or inflammatory responses, and (2) the local microenvironment  ... 
doi:10.1136/gut.2004.044917 pmid:15591496 pmcid:PMC1774348 fatcat:e5k7knipwrfgnd3zfzjos5blb4

SYDNEY CMCRC at TAC 2013

Glen Pink, Andrew Naoum, Will Radford, Will Cannings, Joel Nothman, Daniel Tse, James R. Curran
2013 Text Analysis Conference  
The system extends our TAC 2012 system (Radford et al., 2012) , introducing new features for modelling local entity description and type-specific matching as well type-specific supervised models and supervised  ...  Departing from our previous English NEL submissions (Radford et al., 2010; Radford et al., 2012) , we adopt a supervised approach to disambiguation for NEL.  ...  Data Preprocessing and Resources We continue (see Radford et al., 2012) to link against the Wikipedia dump from April 2012 1 .  ... 
dblp:conf/tac/PinkNRCNTC13 fatcat:e7r2mi2fhzgstmg4h6avnsct6y

Learning to generate one-sentence biographies from Wikidata [article]

Andrew Chisholm, Will Radford, Ben Hachey
2017 arXiv   pre-print
In the future, we will explore whether results improve with explicit modelling of facts and conditioning of generation and autoencoding losses on slots.  ... 
arXiv:1702.06235v1 fatcat:ur5mgi4oo5ebpgaiib3r4qavna

Discriminating between similar languages in Twitter using label propagation [article]

Will Radford, Matthias Galle
2016 arXiv   pre-print
Identifying the language of social media messages is an important first step in linguistic processing. Existing models for Twitter focus on content analysis, which is successful for dissimilar language pairs. We propose a label propagation approach that takes the social graph of tweet authors into account as well as content to better tease apart similar languages. This results in state-of-the-art shared task performance of 76.63%, 1.4% higher than the top system.
arXiv:1607.05408v1 fatcat:j7v7wkjwfra53bdrg4acu2rzlu

(Almost) Total Recall - SYDNEY CMCRC at TAC 2012

Will Radford, Will Cannings, Joel Nothman, Daniel Tse, James R. Curran, Andrew Naoum, Glen Pink
2012 Text Analysis Conference  
The top section shows updated systems while results from the bottom section are drawn from previous system reports (Radford et al., 2010; Radford et al., 2011) .  ...  In analyzing the errors made by our previous English NEL systems (Radford et al., 2010; Radford et al., 2011) , we identified two key areas for improvement: candidate recall and supervised learning.  ... 
dblp:conf/tac/RadfordCNTCNP12 fatcat:t6n6gcdu7bfvfhwrqjaz2suysu

Presenting a New Dataset for the Timeline Generation Problem [article]

Xavier Holt, Will Radford, Ben Hachey
2016 arXiv   pre-print
It is a given that certain entity's will be involved in more newsworthy events than others.  ...  As such, the dataset we derive our reference timelines from must also share this property. • Manageability: Each entity-article pair will be subject to a number of crowd-judgments.  ... 
arXiv:1611.02025v1 fatcat:wpwg3wv74vhyrj7n7sep272hfy

Post-edit Analysis of Collective Biography Generation [article]

Bo Han, Will Radford, Anaïs Cadilhac, Art Harol, Andrew Chisholm, Ben Hachey
2017 arXiv   pre-print
Text generation is increasingly common but often requires manual post-editing where high precision is critical to end users. However, manual editing is expensive so we want to ensure this effort is focused on high-value tasks. And we want to maintain stylistic consistency, a particular challenge in crowd settings. We present a case study, analysing human post-editing in the context of a template-based biography generation system. An edit flow visualisation combined with manual characterisation
more » ... f edits helps identify and prioritise work for improving end-to-end efficiency and accuracy.
arXiv:1702.05821v1 fatcat:dx7tod3iuzeurj767pe3hz6zgi

Dense Molecular Gas in Extreme Starburst Galaxies - What will we learn from Herschel? [article]

T. R. Greve, S. J. E. Radford Institut für Astronomie, ETH, Zürich, Switzerland; Purple Mountain Observatory, Nanjing, China)
2006 arXiv   pre-print
In addition, we have build molecular line templates based on our observations, and demonstrate that Herschel/HI-FI will be able to detect the high-J transitions of most of the above molecules in a large  ...  Radford, Solomon & Downes 1991) and is estimated from the best-fit LVG solution to the HCN line ratios.  ...  Introduction The coming years will witness the emergence of a new generation of groundbased and space-born mm/sub-mm telescopes with sensitivities and frequency coverage that will allow us to study the  ... 
arXiv:astro-ph/0609826v1 fatcat:ecmwmgmfw5agxagfpudauo6rfe

Evaluating Entity Linking with Wikipedia

Ben Hachey, Will Radford, Joel Nothman, Matthew Honnibal, James R. Curran
2013 Artificial Intelligence  
We will follow the terminology of these papers, and refer to the three tasks respectively as cross-document coreference resolution (cdcr), wikification, and named entity linking (nel).  ...  For named entity disambiguation, there is little reason to believe that two people named John Smith will share any more properties than one entity named Paul Simonell and another named Hugh Diamoni, so  ... 
doi:10.1016/j.artint.2012.04.005 fatcat:n7yrkdxdnzat7pk3m2qx4khn7y

Graph-Based Named Entity Linking with Wikipedia [chapter]

Ben Hachey, Will Radford, James R. Curran
2011 Lecture Notes in Computer Science  
The nature and distribution of Wikipedia and WordNet links are very different, and so there is no guarantee WSD approaches will work for NEL.  ...  This assumes that longer mentions (e.g., David Murray) are generally more specific than shorter mentions (e.g., Murray) and will thus be easier to disambiguate.  ... 
doi:10.1007/978-3-642-24434-6_16 fatcat:ekg3op6i2ncc5iq4lvt3ltdtgq

Learning multilingual named entity recognition from Wikipedia

Joel Nothman, Nicky Ringland, Will Radford, Tara Murphy, James R. Curran
2013 Artificial Intelligence  
Radford was supported by an Australian Postgraduate Award. Nothman and Radford were also supported by a Capital Markets CRC PhD top-up scholarship.  ...  Dakka and Cucerzan [19] suggest that most humans will be able to classify an article after reading its first paragraph.  ...  Our results demonstrate this approach will be highly effective and efficient for creating ner models in resource-scarce languages.  ... 
doi:10.1016/j.artint.2012.03.006 fatcat:7agjkau5wfhqbeyit3sddv2ggy

Naïve but effective NIL clustering baselines - CMCRC at TAC 2011

Will Radford, Joel Nothman, James R. Curran, Ben Hachey, Matthew Honnibal
2011 Text Analysis Conference  
These are explained in more detail in Sections 5 and 6 of our notebook paper last year (Radford et al., 2010) .  ...  We first use our best TAC 2010 system (Radford et al., 2010) to link all queries. Any NIL-assigned queries are clustered, creating distinct NIL IDs.  ... 
dblp:conf/tac/RadfordNCHH11 fatcat:pzwasy5qjnaq5kwtk2x66bhmea

Joint Apposition Extraction with Syntactic and Semantic Constraints

Will Radford, James R. Curran
2013 Annual Meeting of the Association for Computational Linguistics  
Our results will immediately help the many systems that already use apposition extraction components, such as coreference resolution and IE.  ...  Our results will immediately benefit the large number of systems with apposition extraction components for coreference resolution and IE.  ... 
dblp:conf/acl/RadfordC13 fatcat:ntnklyobwbavjoyxbk662qxttm

Discovering Entity Knowledge Bases on the Web

Andrew Chisholm, Will Radford, Ben Hachey
2016 Proceedings of the 5th Workshop on Automated Knowledge Base Construction  
Recognition and disambiguation of named entities in text is a knowledge-intensive task. Systems are typically bound by the resources and coverage of a single target knowledge base (KB). In place of a fixed knowledge base, we attempt to infer a set of endpoints which reliably disambiguate entity mentions on the web. We propose a method for discovering web KBs and our preliminary results suggest that web KBs allow linking to entities that can be found on the web, but may not merit a major KB entry.
doi:10.18653/v1/w16-1302 dblp:conf/akbc/ChisholmRH16 fatcat:56tjpxt4tzffljrugcn3xw25ba

:telephone::person::sailboat::whale::okhand:; or "Call me Ishmael" - How do you translate emoji? [article]

Will Radford and Andrew Chisholm and Ben Hachey and Bo Han
2016 arXiv   pre-print
We report on an exploratory analysis of Emoji Dick, a project that leverages crowdsourcing to translate Melville's Moby Dick into emoji. This distinctive use of emoji removes textual context, and leads to a varying translation quality. In this paper, we use statistical word alignment and part-of-speech tagging to explore how people use emoji. Despite these simple methods, we observed differences in token and part-of-speech distributions. Experiments also suggest that semantics are preserved in
more » ... he translation, and repetition is more common in emoji.
arXiv:1611.02027v1 fatcat:h7z5lmlgm5au5i4q4h3aapttjy
« Previous Showing results 1 — 15 out of 37,178 results