ACL RD-TEC 1.0 Summarization of C94-1032
Paper Title:
A STOCHASTIC JAPANESE MORPHOLOGICAL ANALYZER USING A FORWARD-DP BACKWARD-A* N-BEST SEARCH ALGORITHM
A STOCHASTIC JAPANESE MORPHOLOGICAL ANALYZER USING A FORWARD-DP BACKWARD-A* N-BEST SEARCH ALGORITHM
Primarily assigned technology terms:
- algorithm
- analysis method
- analyzer
- bracketing
- character recognition
- character recognition \
- computing
- database
- decoding
- dp search
- dynamic programming
- dynamic programming algorithm
- dynamic programming search
- em algorithm
- forward dynamic programming search
- forward search
- forward-backward algorithm
- information retrieval
- japanese morphological analysis
- japanese morphological analyzer
- kana-to-kanji conversion
- language modeling
- language modeling technique
- learning
- learning method
- machine translation
- matching
- modeling
- modeling technique
- morphological analysis
- morphological analysis \
- morphological analyzer
- morphological analyzers
- morphology
- n-best search
- nlp
- parsers
- part of speech tagging
- programming algorithm
- recognition
- recognition systems
- reporting
- scoring
- search
- search algorithm
- segmentation
- smoothing
- smoothing problem
- speech recognition
- speech recognition \
- speech synthesis
- speech tagging
- spelling
- statistical language modeling
- stochastic japanese morphological analysis \
- stochastic tagger
- supervised learning
- synthesis
- tag assignment
- tagger
- tagging
- tile
- tile tagging
- tree search
- viterbi
- viterbi algorithm
- viterbi algorithm \
- viterbi decoding
- word segmentation
Other assigned terms:
- ambiguity
- approach
- bigram
- bigram model
- boundary marker
- bunsetsu
- character sequence
- character type
- characters
- community
- computational model
- conjugation form
- data structures
- dialogues
- dictionaries
- dictionary
- estimation
- evaluation measures
- events
- experimental results
- fact
- grammar
- heuristic
- heuristics
- hypotheses
- hypothesis
- input string
- joint probability
- language model
- lattice
- linear time
- machine readable dictionaries
- markov models
- measures
- method
- n-best search strategy
- nlp community
- open test
- parse
- parse structure
- part of speech
- part of speech tags
- partial parse
- partial parses
- parts of speech
- precision
- probabilities
- probability
- probability estimates
- proper noun
- relative frequency
- search strategy
- segmentation accuracy
- sentence
- sentence boundaries
- sentence boundary
- sentences
- slot
- speech tag
- statistical approach
- statistical language model
- symbol
- symbols
- tag sequence
- tag set
- tagged text
- tagging model
- tags
- technique
- test set
- text
- text revision support
- tile corpus
- tile system
- tile word
- tim sentence
- tokens
- training
- training corpus
- training data
- tree
- trigram
- trigram model
- two-pass n-best search strategy
- uniform probability
- unigram
- verb
- word
- word boundaries
- word boundary
- word formation
- word information
- word model
- word sequence
- word types
- words
