Tag-assisted sentence confabulation for intelligent text recognition

Autonomous and intelligent recognition of printed or handwritten text image is one of the key features to achieve situational awareness. A neuromorphic model based intelligent text recognition (ITR) system has been developed in our previous work, which recognizes texts based on word level and sentence level context represented by statistical information of characters and words. While quite effective, sometimes the existing ITR system still generates results that are grammatically incorrect because it ignores semantic and syntactic properties of sentences. In this work, we improve the accuracy of the existing ITR system by incorporating parts-of-speech tagging into the text recognition procedure. Our experimental results show that the tag-assisted text recognition improves sentence level success rate by 33% in average.

[1]  Heshaam Faili.  Building deep dependency structure from partial parses , 2009, 2009 14th International CSI Computer Conference.

[2]  Shrikanth S. Narayanan,et al.  A Statistical Approach for Modeling Prosody Features using POS Tags for Emotional Speech Synthesis , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[3]  Adwait Ratnaparkhi,et al.  A Maximum Entropy Model for Part-Of-Speech Tagging , 1996, EMNLP.

[4]  Qing Wu,et al.  Confabulation based sentence completion for machine reading , 2011, 2011 IEEE Symposium on Computational Intelligence, Cognitive Algorithms, Mind, and Brain (CCMB).

[5]  Qing Wu,et al.  Unified perception-prediction model for context aware text recognition on a heterogeneous many-core platform , 2011, The 2011 International Joint Conference on Neural Networks.

[6]  Jerome R. Bellegarda,et al.  Improved pos tagging for text-to-speech synthesis , 2011, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[7]  Patrick van der Smagt,et al.  Introduction to neural networks , 1995, The Lancet.

[8]  Christopher D. Manning,et al.  Enriching the Knowledge Sources Used in a Maximum Entropy Part-of-Speech Tagger , 2000, EMNLP.

[9]  Qing Wu,et al.  A Parallel Neuromorphic Text Recognition System and Its Implementation on a Heterogeneous High-Performance Computing Cluster , 2013, IEEE Transactions on Computers.

[10]  Hermann Ney,et al.  Does the Cost Function Matter in Bayes Decision Rule? , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Srinivas Bangalore,et al.  Complexity of lexical descriptions and its relevance to partial parsing , 1997 .

[12]  Robert Hecht-Nielsen.  Confabulation theory - the mechanism of thought , 2007 .

[13]  Zhong Liu,et al.  A Novel Method of Chinese Web Information Extraction and Applications , 2009, 2009 WASE International Conference on Information Engineering.