Analysis of Distributed English Chunk Recognition under Correlation Evaluation Based on Deep Belief Network Model

Most of the traditional methods English text chunk recognition are solved by setting corresponding phrase identifier numbers and eventually transforming problem into a lexical annotation problem. In language recognition, MFCC features easily contaminated noise have weak immunity due to insufficient amount information on each frame signal. At same time, SDC feature extraction commonly used today require artificial settings in parameter selection, which increases uncertainty results. The method identifying chunks association evaluation central word extensions identifies from different perspective. It has following features: (i) phase is considered as cluster with core, internal composition pattern fully considered; (ii) results dynamically evaluated using confidence. show that proposed can achieve higher rate than methods. faster, F -measure value block reaches 94.05%, comparable best so far.