This work aimed to adopt a transformer model combined with deep learning neural network to discuss the segmentation of medical ultrasound images. A network combining a transformer model with a deep neural network model (ConvTrans-Net) is proposed. The image content is preselected based on a multilayer perceptron and attention mechanism, different feature vectors are concatenated and fed into the multilayer perceptron, and the results of multiple attentions are mapped to a larger dimensional space using a feed-forward network. The lesion areas segmented by ultrasonic scan were analysed, and an attention mechanism and multilayer perceptron were combined to preselect image content. The performance and convergence of the model were analysed, and the Jaccard similarity coefficient precision and recall of the model were measured. In the experiment, two different iterative step sizes were selected, the convergence trend of the model increased with the increase in the number of iterative steps, and the model gradually stabilized. The Jaccard of ConvTrans-Net was 85.21%. The precision (85.17%) and recall (89.65%) were significantly higher than those of EfficientNet and DeepViT-L, and the differences were significant (P < 0.05). The experimental results show that the proposed model is stable, and the combination of Transformer model and deep learning neural network has a good effect on ultrasound image segmentation, which has some practical application value.
The purpose of sentiment classification is to solve the problem of automatic judgment of text sentiment tendency. In the sentiment classification task of online reviews, traditional deep learning sentiment classification models focus on algorithm optimization to improve the classification performance of the model, but when the sample data for manually labeling sentiment tendencies is insufficient, the classification performance of the model will be poor. The deep learning sentiment classification model based on weak tagging information, on the one hand, introduces weak tagging information into the training process of the model to reduce the use of manually tagging data. On the other hand, weak tagging information can represent the sentiment tendency of reviews to a certain extent, but it also contains noise, the model reduces the negative impact of the noise in weak tagging information in order to improve the classification performance of the sentiment classification model. The experimental results show that in the sentiment classification task of hotel online reviews, the deep learning sentiment classification model based on weak tagging information has superior classification performance than the traditional deep model without increasing labor cost.
9 September 2016 in Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
Business Intelligence System (BIS) has become an important tool for enterprises to make decision timely and effectively. However there are many differences in the quality and performance of the BIS on the market, it is necessary for enterprise manage