In the field of military research, manufacturing and management of weapons and equipment are very important. Due to the continuous advancement of science and technology, many military equipment databases have a loose structure, which makes them difficult to be utilized efficiently, resulting in low efficiency, chaotic management, and other issues. In order to solve these problems, an entity-relation extraction method based on CRF and syntactic analysis tree is proposed according to the latest text extraction algorithm. Finally, a military knowledge graph construction method is optimized via massive data training, model comparison and improvement. The ternary data extraction method is significantly better than the single algorithm extraction method, and the accuracy of the extracted training model can reach 72%. Compared with the traditional entity-relation extraction method, the accuracy of the entityrelation extraction method based on the fusion of CRF and syntax analysis tree is improved by 12.6% when the confidence model is added, and the comprehensive evaluation accuracy can reach 78.11%. This result has significant practical value for the construction of knowledge graphs in the field of military equipment.
Facial expression recognition technology has become a powerful tool for conveying human emotions and intentions and is widely used in areas such as assisted driving and intelligent medical care. Due to the limited computing power of current hardware devices and the real‐time requirements of application scenarios, this paper proposes a high‐performance and lightweight framework for real‐time facial expression recognition framework to solve the problem of real‐time completion of expression recognition tasks under low hardware costs. To address these issues, this paper first designs a RepVGG and mobileNetV2 dual‐channel structure in the feature extraction. It is then input into the MobileViT Block for global feature modelling. Finally, the position vector of the capsule network is used to replace the output of the global pooling, preserving the spatial relationship of the salient features and enhancing the classification effect. Compared with the mainstream facial expression recognition algorithm that cannot get good classification results under low complexity conditions, the model has a significant accuracy improvement while ensuring lightweight. With only 294.60M FLOPS and 0.95M parameters, it achieved an accuracy of 97.53% on the KDEF dataset and 85.56% on the RAF‐DB, demonstrating the advanced nature of the algorithm.
NER (Named Entity Recognition) is of great significance for the construction of a knowledge map. The purpose is to guarantee the recognition effect of named entity recognition method in the application scenario of vertical field, a named entity recognition method is proposed based on BI-LSTM-CRF [BI(Bidirectional) LSTM (Long-Short Term Memory) CRF (Conditional Random Field)] for equipment support field, which improves the recognition effect of the domain named entity and provides technical support for the subsequent construction of domain knowledge map. First, Chinese characters are represented by word embedding and input into the model. Then, the input feature vector sequence is processed through BI-LSTM NN (Neural Network) to extract contextual semantic learning features. Finally, the learned features are connected to the linear CRF, the NEs (Named Entity) in the field of equipment support are labeled, and the NER results are obtained and output. The experimental results show that the precision of the named entity recognition method based on the BI-LSTM-CRF model has reached 92.02%, the recall rate has reached 93.21%, and the F1score has reached 93.88%. Meanwhile, the performance of the proposed BI-LSTM-CRF model is higher than the precision of the BI-LSTM NN model and LSTM-CRF NN model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.