... with facility for rich user interaction.
Our starting in the development of this system
was the Ushahidi platform,
1
which has high up-
take for social media surveillance and information
dissemination ... information
dissemination purposes across a range of organ-
isations. The reason for us choosing to imple-
ment our own platform was: (a) ease of integra-
tion of back-end...
... Cohn
∗
tcohn@inf.ed.ac.uk
Miles Osborne
∗
miles@inf.ed.ac.uk
∗
Department of Informatics
University of Edinburgh
Edinburgh, EH8 9AB, UK
†
Department of Linguistics
University of Maryland
College Park, MD 20742, USA
Abstract
We ... model per-
forms well on languages for which the monotone
bias of existing alignment and heuristic phrase
extraction approaches fail. These results open...
... and information-based methods, e.g.
pointwise mutual information (Church & Hanks
1990).
Several studies of the performance of lexical
association metrics suggest significant room for
improvement, ... general level of performance was low,
with precision falling off rapidly as larger portions
605
of the n-best list were included, but they report
better performance with sta...
... list of sentences as
hyperlinks to the source documents. This is conve-
nient for Web users but should be transformed for
spoken output. It now offers plain text as an alterna-
tive to HTML for ... systems
for the Pocket PC that interpret queries spoken in
English or Chinese. This last group appears to be at
the forefront of current research in spoken interfaces
for document ret...
... Major
League Baseball) in 2009. Due to the large size of
the corpora, we uniformly sampled a subset of doc-
uments for each corpus and ran the Stanford NER
tagger (Finkel et al., 2005), which tagged ... here consists of
identifying the correct team for every player. The
EEA model is not designed for this and performed
poorly. Our model can do better, since it makes use
of cont...
... Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics, pages 330–338,
Jeju, Republic of Korea, 8-14 July 2012.
c
2012 Association for Computational Linguistics
A ... construction of a novel Bayesian framework for
semantic parsing of natural language (NL) utter-
ances in a unifying framework in §4,
(ii) representation of seed labeled data an...
... high-level
expression of Ab peptides should lead to aggregation
and formation of inclusion bodies, and that Ab would
be less susceptible to degradation in this form. More-
over, the formation of inclusion ... mean of the fluorescence signal of at least eight wells
containing aliquots of the same solution. The morphology
of Ab aggregates formed from solutions incubated as above
bu...
... symbol “q”. A sequence of multigrams forms
an alignment of a source and target word. The prob-
ability of a sequence of multigrams a is the product
of the probabilities of the multigrams it contains.
p(a) ... returns the set of all valid alignment se-
quences a of a word pair (e, f ). The joint transliter-
ation probability p
1
(e, f ) of a word pair is the sum
of the prob...
... effort, yet there is very limited avail-
ability of off-the shelf tools for researchers
whose interests are not in coreference, or for
researchers who want to concentrate on a
specific aspect of ... of MMAX2’s functions. Using a
generic format for standoff annotation allows the use
of the coreference resolution as part of a larger sys-
tem, but also performing qualitative error...