Constitutive Artificial Neural Network for the Construction of an English Multimodal Corpus

Multimodal corpus is a novel multimedia teaching tool in social development and educational reform process. It uses a range of multimedia components to build a wide-ranging English corpus and is mostly focused on computer and network technologies. The use of corpus in multimodal English-Chinese instruction is growing. The meaning, usage, and set of English and Chinese multimodality will be better understood with the aid of contemporary information technology, which will also enhance the initiative of autonomous learning. This paper builds a Multimodal English Corpus created on optimized-Constitutive Artificial Neural Network-Honey Badger Algorithm (MEC-CANN-HBA). The input data is collected via the dataset of Gutenberg Literary English Corpus (GLEC). The data are fed to pre-processing to remove the noise and enhance the input data utilizing Multivariate Fast Iterative Filtering (MFIF). The pre-processing output is given to the Feature extraction segment. The three significant features, such as text, audio and video are extracted based on Deep Wavelet Scattering Transform (DWST). After that, the extracted features are given to the multimodal fusion vector. The multimodal feature vectors are employed as the input data for categorization and to obtain the English poetry feature representation that integrates context characteristics. Finally, the features of output are used as the input data of Constitutive Artificial Neural Network (CANN) effectively categorizes as ideographic, phonetic, rhetorical and contextual. Honey Badger Algorithm (HBA) utilized for improving the weight parameter of CANN to check the classification of English poetry is current utterance. The proposed MEC-CANN-HBA approach attains 24.36%, 23.42%, 30%, 10.25% and 16.27% higher accuracy, and 26.61%, 28.50%, 23%, 18.33% and 21.24% greater precision rate, compared with existing methods, like Construction for Multiple Modal Corpus of College Students’ Spoken English Using Semantic Concepts (CMC-CSSE-SC), Construction of multimodal poetry translation corpus under AdaBoost method (CMPTC- AdaBoost),Construction with Application of English-Chinese Multimodal Emotional Corpus utilizing Artificial Intelligence (CECMC-AI) respectively