Common Sense Knowledge for Handwritten Chinese Text Recognition

Qiu Feng Wang*, Erik Cambria, Cheng Lin Liu, Amir Hussain

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

67 Citations (Scopus)

Abstract

Compared to human intelligence, computers are far short of common sense knowledge which people normally acquire during the formative years of their lives. This paper investigates the effects of employing common sense knowledge as a new linguistic context in handwritten Chinese text recognition. Three methods are introduced to supplement the standard n-gram language model: embedding model, direct model, and an ensemble of these two. The embedding model uses semantic similarities from common sense knowledge to make the n-gram probabilities estimation more reliable, especially for the unseen n-grams in the training text corpus. The direct model, in turn, considers the linguistic context of the whole document to make up for the short context limit of the n-gram model. The three models are evaluated on a large unconstrained handwriting database, CASIA-HWDB, and the results show that the adoption of common sense knowledge yields improvements in recognition performance, despite the reduced concept list hereby employed.

Original languageEnglish
Pages (from-to)234-242
Number of pages9
JournalCognitive Computation
Volume5
Issue number2
DOIs
Publication statusPublished - Jun 2013
Externally publishedYes

Keywords

  • Common sense knowledge
  • Handwritten Chinese text recognition
  • Linguistic context
  • Natural language processing
  • n-gram

Cite this