Emotion recognition plays numerous important roles in the life of individuals in the context of artificial intelligence technology. The majority of existing emotion recognition techniques performs poorly in practical applications, preventing their advancement. Hence, put forth a Deep Kronecker Neural Network based multimodal expression EEG interaction technique to address this issue. First, Hexadecimal Local Adaptive Binary Pattern (HLABP) is used as an objective way of feature extraction. Depending on facial expression, the features are selectedwith the help of Weibull Distributive Generalized Multidimensional Scaling (WDGMS), the solution vector coefficients are scrutinized to scale the facial expression type of test samples. Finally, Deep Kronecker Neural Network (DKNN)completesthe classification task. Then, the proposed method is simulated utilizing MATLAB under several performance metrics, like F1 score, accuracy, error rate, average running time. The proposed technique attains 23.34%, 16.64% higher accuracy and 34.61%, 41.23% lower average running time when comparing to the existing methods, such as Expression-EEG Interaction Multiple Modal Emotion Recognition utilizing Deep Automatic Encoder (EEG-MER-DAE) and EEG-based emotion recognition utilizing deep convolutional neural networks (EEG-DNN) respectively.