The proposed emotion recognition model is based on the hierarchical long-short term memory neural network (LSTM) for video-electroencephalogram (Video-EEG) signal interaction. The inputs are facial-video and EEG signals from the subjects when they are watching the emotion-stimulated video. The outputs are the corresponding emotion recognition results. Facial-video features and corresponding EEG features are extracted based on a fully connected neural network (FC) at each time point. These features are fused through hierarchical LSTM to predict the key emotional signal frames at the next time point until the emotion recognition result is calculated at the last time point. Specially, a self-attention mechanism is applied to show the correlation of the stacked LSTM at different hierarchies. In this process, the “selective focus” is used to analyze the human-emotional temporal sequences in each model, which improves the utilization of the key spatial EEG signals. Moreover, the process includes the temporal attention mechanism to predict the key signal frame at next time point, which utilizes the key emotion data in temporal domain. The experimental results prove that the classification rate (CR) and F1-score of the proposed emotion recognition model are significantly increased by at least 2% and 0.015, respectively, compared to other methods.

In recent research, supervised image clustering based on Graph Neural Networks (GNN) connectivity prediction has demonstrated considerable improvements over traditional clustering algorithms. However, existing supervised image clustering algorithms are usually time-consuming and limit their applications. In order to infer the connectivity between image instances, they usually created a subgraph for each image instance. Due to the creation and process of a large number of subgraphs as the input of GNN, the computation overheads are enormous. To address the high computation overhead problem in the GNN connectivity prediction, we present a time-efficient and effective GNN-based supervised clustering framework based on density division namely DDC-GNN. DDC-GNN divides all image instances into high-density parts and low-density parts, and only performs GNN subgraph connectivity prediction on the low-density parts, resulting in a significant reduction in redundant calculations. We test two typical models in the GNN connectivity prediction module in the DDC-GNN framework, which are the graph convolutional networks (GCN)-based model and the graph auto-encoder (GAE)-based model. Meanwhile, adaptive subgraphs are generated to ensure sufficient contextual information extraction for low-density parts instead of the fixed-size subgraphs. According to the experiments on different datasets, DDC-GNN achieves higher accuracy and is almost five times quicker than those without the density division strategy.

In order to solve the problem that the existing deep learning method has insufficient ability in feature extraction in the text emotion classification task, this paper proposes a text emotion analysis using the dual-channel convolution neural network in the social network. First, a double-channel convolutional neural network is constructed. Combined with emotion words, parts of speech, degree adverbs, negative words, punctuation, and other word features that affect the texts emotional tendency, an extended text feature is formed. Then, using the CNNs multichannel mechanism, the extended text features based on the word vector features and the semantic features based on the word vectors are, respectively, input into the CNN model. After each convolution operation of the convolution channel, the BN technology is used to normalize the internal data of the network and the padding technology is used to improve the ability of the model to extract edge features of the data and the speed of the model. Finally, a dynamic k-max continuous pooling strategy is adopted to realize the dimensionality reduction of features and enhance the models ability to extract features. The experimental results show that the accuracy and F1 values obtained by the proposed method can be as high as 94.16 and 92.61, respectively, which are better than several comparison algorithms.