ACL RD-TEC 1.0 Summarization of P06-1126

Paper Title:
DISCRIMINATIVE PRUNING OF LANGUAGE MODELS FOR CHINESE WORD SEGMENTATION

Authors: Jianfeng Li and Haifeng Wang and Dengjun Ren and Guohua Li

Other assigned terms:

  • approach
  • association for computational linguistics
  • backoff
  • bigram
  • bigram language model
  • bigram model
  • case
  • characters
  • chinese characters
  • chinese language
  • chinese word
  • coefficient
  • conditional probabilities
  • correlation
  • correlation coefficient
  • correlations
  • document
  • entropy
  • evaluation metrics
  • experimental results
  • f-measure
  • fact
  • generative model
  • gold standard
  • knowledge
  • kullback-leibler distance
  • language model
  • language model perplexity
  • language models
  • language processing applications
  • language processing tasks
  • likelihood
  • linguistics
  • measure
  • measures
  • method
  • model perplexity
  • model size
  • n-gram
  • n-gram language model
  • n-grams
  • natural language
  • natural language processing applications
  • perplexity
  • probabilities
  • probability
  • process
  • processing tasks
  • segmentation bakeoff
  • segmented corpus
  • sentence
  • sentences
  • source-channel model
  • statistical models
  • system performance
  • terms
  • text
  • toolkit
  • training
  • training corpus
  • training data
  • unigram
  • unigram model
  • unigram probability
  • vocabulary
  • word
  • word segmentation performance
  • word sequence
  • word sequences
  • words

Extracted Section Types:


This page last edited on 10 May 2017.

*** ***