Learn More
There has been much interest in using phrasal movement to improve statistical machine translation. We explore how well phrases cohere across two languages, specifically English and French, and examine the particular conditions under which they do not. We demonstrate that while there are cases where coherence is poor, there are many regularities which can be(More)
Since 1995, a few statistical parsing algorithms have demonstrated a breakthrough in parsing accuracy, as measured against the UPenn TREEBANK as a gold standard. In this paper we report adapting a lexicalized, probabilistic context-free parser to information extraction and evaluate this new technique on MUC-7 template elements and template relations.
For MUC-7, BBN has for the first time fielded a fully-trained system for NE, TE, and TR; results are all the output of statistical language models trained on annotated data, rather than programs executing handwritten rules. Such trained systems have some significant advantages: • They can be easily ported to new domains by simply annotating data with(More)
We describe a series of three experiments in which supervised learning techniques were used to acquire three different types of grammars for English news stories. The acquired grammar types were: 1) context-free, 2) context-dependent, and 3) probabilistic context-free. Training data were derived from University of Pennsylvania Treebank parses of 50 Wall(More)
All of BBN's research under the TIPSTER III program has focused on doing extraction by applying statistical models trained on annotated data, rather than by using programs that execute hand-written rules. Within the context of MUC7, the SIFT system for extraction of template entities (TE) and template relations (TR) used a novel, integrated(More)
One hundred and forty-two consecutive patients with proximal femoral fractures were audited prospectively over a 1-year period. Mobility, age and sex were recorded along with timing of surgery, complications, 'will to live', length of admission, mortality, mobility and housing requirements on discharge. Operative procedures were performed mostly by(More)
Nineteen patients with Duchenne's muscular dystrophy underwent segmental spinal instrumentation and posterior fusion between 1989 and 1994. The indication for surgery was loss of the ability to walk and development of scoliosis with sitting discomfort. Preoperative assessment included evaluation of pulmonary function. Average age at operation was 12.5(More)
Length of hospital stay and outcome after femoral neck fracture were compared in a prospective study between two adjacent hospitals. In matched populations, mean length of stay was 30.8 days at Hospital X and 15.7 days at Hospital Y. Need for rehousing, age over 80 years and new nursing home placement prolonged length of stay at Hospital X, but not at(More)
Traditional approaches to the problem of extracting data from texts have emphasized hand-crafted linguisti c knowledge . In contrast, BBN's PLUM system (Probabilistic Language Understanding Model) was developed a s part of a DARPA-funded research effort on integrating probabilistic language models with more traditional linguistic techniques . Our research(More)