#14279However, a great deal of natural language texts e.g., memos, rough drafts,conversation transcripts etc., have features that differ significantly from neat texts, posing special problems for readers, such as misspelled words, missing words, poor syntactic construction, missing periods, etc.
other,23-1-P82-1035,ak
be in reasonably neat form , e.g. ,
<term>
newspaper stories
</term>
and other edited texts . However
#14256Most large text-understanding systems have been designed under the assumption that the input text will be in reasonably neat form, e.g.,newspaper stories and other edited texts.
other,26-4-P82-1035,ak
with multiple
<term>
meanings
</term>
(
<term>
ambiguity
</term>
) , fill in missing words (
<term>
#14370These syntactic and semantic expectations can be used to figure out unknown words from context, constrain the possible word-senses of words with multiple meanings (ambiguity), fill in missing words (ellipsis), and resolve referents (anaphora).
other,41-4-P82-1035,ak
and resolve
<term>
referents
</term>
(
<term>
anaphora
</term>
) . This method of using expectations
#14385These syntactic and semantic expectations can be used to figure out unknown words from context, constrain the possible word-senses of words with multiple meanings (ambiguity), fill in missing words (ellipsis), and resolve referents (anaphora).
other,34-4-P82-1035,ak
</term>
) , fill in missing words (
<term>
ellipsis
</term>
) , and resolve
<term>
referents
</term>
#14378These syntactic and semantic expectations can be used to figure out unknown words from context, constrain the possible word-senses of words with multiple meanings (ambiguity), fill in missing words (ellipsis), and resolve referents (anaphora).
tool,21-5-P82-1035,ak
a working computer program called
<term>
NOMAD
</term>
, which understands scruffy
<term>
#14409This method of using expectations to aid the understanding of scruffy texts has been incorporated into a working computer program calledNOMAD, which understands scruffy texts in the domain of Navy messages.
other,14-4-P82-1035,ak
used to figure out unknown words from
<term>
context
</term>
, constrain the possible
<term>
word-senses
#14358These syntactic and semantic expectations can be used to figure out unknown words fromcontext, constrain the possible word-senses of words with multiple meanings (ambiguity), fill in missing words (ellipsis), and resolve referents (anaphora).
tech,2-1-P82-1035,ak
executed to yield the answer . Most large
<term>
text-understanding systems
</term>
have been designed under the assumption
#14235Most largetext-understanding systems have been designed under the assumption that the input text will be in reasonably neat form, e.g., newspaper stories and other edited texts.
other,41-2-P82-1035,ak
misspelled
<term>
words
</term>
, missing
<term>
words
</term>
, poor
<term>
syntactic construction
#14304However, a great deal of natural language texts e.g., memos, rough drafts, conversation transcripts etc., have features that differ significantly from neat texts, posing special problems for readers, such as misspelled words, missingwords, poor syntactic construction, missing periods, etc.
other,38-2-P82-1035,ak
problems for readers , such as misspelled
<term>
words
</term>
, missing
<term>
words
</term>
, poor
#14301However, a great deal of natural language texts e.g., memos, rough drafts, conversation transcripts etc., have features that differ significantly from neat texts, posing special problems for readers, such as misspelledwords, missing words, poor syntactic construction, missing periods, etc.
other,24-4-P82-1035,ak
</term>
of
<term>
words
</term>
with multiple
<term>
meanings
</term>
(
<term>
ambiguity
</term>
) , fill in
#14368These syntactic and semantic expectations can be used to figure out unknown words from context, constrain the possible word-senses of words with multiplemeanings (ambiguity), fill in missing words (ellipsis), and resolve referents (anaphora).
other,27-2-P82-1035,ak
that differ significantly from neat
<term>
texts
</term>
, posing special problems for readers
#14290However, a great deal of natural language texts e.g., memos, rough drafts, conversation transcripts etc., have features that differ significantly from neattexts, posing special problems for readers, such as misspelled words, missing words, poor syntactic construction, missing periods, etc.
other,6-2-P82-1035,ak
texts . However , a great deal of
<term>
natural language texts
</term>
e.g. , memos , rough drafts ,
<term>
#14269However, a great deal ofnatural language texts e.g., memos, rough drafts, conversation transcripts etc., have features that differ significantly from neat texts, posing special problems for readers, such as misspelled words, missing words, poor syntactic construction, missing periods, etc.
other,10-3-P82-1035,ak
to these problems is to make use of
<term>
expectations
</term>
, based both on knowledge of surface
#14325Our solution to these problems is to make use ofexpectations, based both on knowledge of surface English and on world knowledge of the situation being described.
other,21-4-P82-1035,ak
possible
<term>
word-senses
</term>
of
<term>
words
</term>
with multiple
<term>
meanings
</term>
#14365These syntactic and semantic expectations can be used to figure out unknown words from context, constrain the possible word-senses ofwords with multiple meanings (ambiguity), fill in missing words (ellipsis), and resolve referents (anaphora).
other,21-3-P82-1035,ak
knowledge of surface English and on
<term>
world knowledge
</term>
of the situation being described
#14336Our solution to these problems is to make use of expectations, based both on knowledge of surface English and onworld knowledge of the situation being described.
other,44-2-P82-1035,ak
</term>
, missing
<term>
words
</term>
, poor
<term>
syntactic construction
</term>
, missing periods , etc . Our solution
#14307However, a great deal of natural language texts e.g., memos, rough drafts, conversation transcripts etc., have features that differ significantly from neat texts, posing special problems for readers, such as misspelled words, missing words, poorsyntactic construction, missing periods, etc.
other,19-4-P82-1035,ak
context
</term>
, constrain the possible
<term>
word-senses
</term>
of
<term>
words
</term>
with multiple
#14363These syntactic and semantic expectations can be used to figure out unknown words from context, constrain the possibleword-senses of words with multiple meanings (ambiguity), fill in missing words (ellipsis), and resolve referents (anaphora).
other,39-4-P82-1035,ak
<term>
ellipsis
</term>
) , and resolve
<term>
referents
</term>
(
<term>
anaphora
</term>
) . This method
#14383These syntactic and semantic expectations can be used to figure out unknown words from context, constrain the possible word-senses of words with multiple meanings (ambiguity), fill in missing words (ellipsis), and resolvereferents (anaphora).
other,11-5-P82-1035,ak
to aid the understanding of scruffy
<term>
texts
</term>
has been incorporated into a working
#14399This method of using expectations to aid the understanding of scruffytexts has been incorporated into a working computer program called NOMAD, which understands scruffy texts in the domain of Navy messages.