Cognitive computing on Chinese Sign Language perception and comprehension

Abstract

After a systematic review on psychological and neurophysiological studies, we propose a computational cognitive model for Chinese Sign Language (CSL) perception and comprehension with detailed algorithmic descriptions based on cognitive functionalities in human language processing. A semantic neural network based on this model is introduced as the knowledge representation method in CSL comprehension. In line with the actual attention of the deaf, a revised game theory is presented by assigning attention. This paper illustrates the applications of the proposed model in classifier predicative comprehension of CSL. After the spreading activation process, each handshape node is assigned with appropriate values that depend on the relationships nodes. Our experimental results demonstrate that the proposed model can effectively improve the performance of handshape associative memory.