ACL RD-TEC 1.0 Summarization of H05-1025
Paper Title:
PREDICTING SENTENCES USING N-GRAM LANGUAGE MODELS
PREDICTING SENTENCES USING N-GRAM LANGUAGE MODELS
Authors: Steffen Bickel and Peter Haider and Tobias Scheffer
Primarily assigned technology terms:
- algorithm
- assistance system
- backtracking
- beam search
- beam search algorithm
- computational linguistics
- computing
- decoder
- decoding
- editing
- em algorithm
- human language
- human language technology
- index-based retrieval
- indexing
- information retrieval
- instance-based learning
- internet
- language processing
- language technology
- learning
- learning method
- linear interpolation
- n-gram completion
- n-gram-based completion
- natural language processing
- personal computer
- predictor
- processing
- reading
- search
- search algorithm
- sentence completion
- translation systems
- translator
- translators
- tuning
- viterbi
- viterbi algorithm
- viterbi beam
- viterbi beam search
- viterbi decoding
- viterbi search
- word prediction
Other assigned terms:
- approach
- association for computational linguistics
- beam
- break
- case
- characters
- confidence measure
- context information
- corpora
- cosine similarity
- data set
- device
- distribution
- document
- document collection
- document collections
- editing assistance function
- empirical results
- entropy
- evaluation metric
- generation
- hypothesis
- implementation
- interpolation
- joint probability
- language model
- language models
- linguistics
- markov models
- measure
- method
- model probability
- n-gram
- n-gram language model
- n-gram model
- n-gram models
- natural language
- precision
- prior probability
- probability
- process
- queries
- recursion
- recursive structure
- relation
- search space
- sentence
- sentence fragment
- sentences
- set size
- technology
- terms
- test collection
- text
- text collection
- training
- training data
- training set
- training set size
- user
- user behavior
- vocabulary
- vocabulary size
- word
- word n-gram model
- word sequence
- words
