ACL RD-TEC 1.0 Summarization of W98-1205
Paper Title:
LOOK-BACK AND LOOK-AHEAD IN THE CONVERSION OF HIDDEN MARKOV MODELS INTO FINITE STATE TRANSDUCERS
LOOK-BACK AND LOOK-AHEAD IN THE CONVERSION OF HIDDEN MARKOV MODELS INTO FINITE STATE TRANSDUCERS
Primarily assigned technology terms:
- algorithm
- approximation
- computational natural language learning
- disambiguation
- encoding
- finite state
- finite state transducer
- finite state transducers
- finite-state transducer
- hidden markov
- hidden markov model
- hidden markov models
- hmms
- language learning
- language processing
- learning
- markov model
- natural language learning
- parsing
- part-of-speech tagging
- phrase extraction
- probability estimation
- processing
- reading
- regular expression
- tagger
- taggers
- tagging
- text analysis
- transducer
- transducers
- viterbi
- viterbi algorithm
Other assigned terms:
- alphabet
- ambiguity
- approach
- case
- composition
- corpora
- empty string
- english corpus
- estimation
- joint probability
- lexicon
- long-distance dependencies
- lookahead
- markov models
- method
- natural language
- negation
- noun phrases
- part-of-speech
- phrase
- probabilities
- probability
- relation
- sentence
- sentence boundaries
- sentence boundary
- sentences
- stochastic model
- surface form
- symbol
- symbols
- tag sequence
- tag set
- tagger model
- tagging accuracy
- tags
- term
- text
- training
- training corpus
- transition probabilities
- tree
- untagged corpus
- verb
- word
- word sequence
- words
