Search

Multi-interest sequential recommendations leverage users' historical behavior to provide recommendations that match multiple interests. Most of these methods have not fully extracted higher-order information hidden in users' interactions and have overlooked the multiple features of items. To this end, this paper proposes a multi-interest model called "multi-interest sequential recommendation with simplified graph convolution and item multi-features (SGCMF)". Firstly, a simplified graph convolution module is designed based on bipartite graphs, which utilizes mean pooling to aggregate neighboring information and employs a feedforward neural network (FNN) for nonlinear transformations and combinations. This method reduces redundant information and captures higher-order relationships, thereby simplifying the complexity of modeling high-order interactions and improving prediction accuracy. Secondly, an item multi-feature extraction module is proposed, which represents item features with multiple vectors, and analyzes each feature from multiple perspectives while preserving important relationships between features. The model correlates multiple features of the item with user interests, thereby achieving a fine-grained analysis of user interests. Extensive experiments are conducted on five real-world scenarios, and the results are compared with state-of-the-art methods. The experimental results show that SGCMF outperforms other baselines. [ABSTRACT FROM AUTHOR]

Facial expression recognition methods based on convolutional neural networks have greatly improved the recognition accuracy. However, the feature information of face images has not been fully extracted in these methods, which are mainly due to insufficient features of facial expressions in some Key positions, such as eyes, nose, and mouth. Adjusting the weights of these key positions can improve the accuracy of facial expression recognition. To address this problem, we propose a differential weight-based multi-dimension and multi-scale HOG algorithm (MDS-HOG). It can extract various types of deep features, thus enriching facial expression feature information. First, facial expression data is preprocessed using ENM region acquisition. Then, on the basis of the original HOG algorithm, the diagonal gradient information of the image is added to obtain the multi-dimension HOG feature. Furthermore, a multi-scale spatial pyramid is constructed to extract multi-scale HOG features. Finally, according to the different contributions of different face regions, a differential weight EMN algorithm is proposed to deal with the weight distribution of face sub-regions. Experimental results demonstrate that the proposed algorithm exhibits superior performance compared with the original and other improved HOG algorithms on JAFFE and CK+ datasets. [ABSTRACT FROM AUTHOR]