Abstract:Based on the existing optimization neural network algorithm, this paper introduces a simple and computationally efficient adaptive mechanism (adaptive exponential decay rate). By applying the adaptive mechanism to the Adadelta algorithm, it can be seen that AEDR-Adadelta acquires the learning rate dynamically and adaptively. At the same time, by proposing an adaptive exponential decay rate, the number and method of configuring hyperparameters can be reduced, and different learning rates can be effectively obtained for different parameters. The model is based on the encoder-decoder structure and adopts a dual-encoder structure. The transformer encoder is used to extract the context information of the sentence; the Bi-GRU encoder is used to extract the information of the source sentence; and the gated structure is used at the decoder side. The input information is integrated, and each part is matched with different attention mechanisms, which improves the model's ability to extract and analyze relevant features in sentences. In order to accurately capture the coherence features in English texts, an improved subgraph matching algorithm is used to mine frequently occurring subgraph patterns in sentence semantic graphs, which are used to simulate the unique coherence patterns in English texts, and then analyze the overall coherence of English texts. According to the frequency of occurrence of different subgraph patterns in the sentence semantic graph, the subgraphs are filtered to generate frequent subgraph sets, and the subgraph frequency of each frequent subgraph is calculated separately. The overall coherence quality of English text is quantitatively analyzed by extracting the distribution characteristics of frequent subgraphs and the semantic values of subgraphs in the sentence semantic graph. According to the experimental results, the algorithm using the adaptive mechanism can reduce the error of the training set and the test set, improve the classification accuracy to a certain extent, and has a faster convergence speed and better text generalization ability. The semantic coherence diagnosis model of English text in this paper performs well in various tasks and has a good effect on improving the automatic correction system of English composition and providing reference for English teachers' composition correction.

Neural Networks for Coordination Analysis

Text Coherence Analysis Based on Deep Neural Network.

Correspondence Analysis Using Neural Networks

Temporal Nuances of Coordination Network Semantics

Automatic Identification of Chinese Coordination Relations

NeuralAC: Learning Cooperation and Competition Effects for Match Outcome Prediction.

Graph Neural Networks for Learning Robot Team Coordination

Automatic Identification of Coordination with Overt Conjunctions Based on Conditional Random Fields

Adaptive modeling and inference of higher-order coordination in neuronal assemblies: A dynamic greedy estimation approach

Meta Neural Coordination

Coordination as inference in multi-agent reinforcement learning

A Novel Computational and Modeling Foundation for Automatic Coherence Assessment

A Deep Learning Model for Correlation Analysis between Electroencephalography Signal and Speech Stimuli

A Novel Coordination Mechanism for Connected and Automated Vehicles in the Multi-Intersection Road Network

Neural Net Models for Open-Domain Discourse Coherence

Application of Adaptive Neural Network Algorithm Model in English Text Analysis

Context-Aware Sparse Deep Coordination Graphs

Neural Network-based Constrained Optimal Coordination for Heterogeneous Uncertain Nonlinear Multi-agent Systems

Investigating the Role of Centering Theory in the Context of Neural Coreference Resolution Systems

The Cooperative Network Architecture: Learning Structured Networks as Representation of Sensory Patterns

An Empirical Evaluation of Neural and Neuro-symbolic Approaches to Real-time Multimodal Complex Event Detection