Relation extraction (RE) is a crucial step for knowledge graph construction, which aims to extract meaningful relations between entity pairs in plain texts. Very few works have been studied on Chinese relation extraction (CRE) in the military field. Moreover, recent deep neural network-based methods have achieved considerable performance but still suffer from three inherent limitations, including overlapping of entities, imbalanced data and the ambiguity. Therefore, this work investigates a novel Multi-Grained Lattice Transformer (MGLT), which leverages external information of lexicon and word sense tailored for CRE. In MGLT, self-matched lexicon words and related word senses are fused through a cross-transformer mechanism to alleviate the ambiguity in texts. The finally enriched sequence representation in MGLT captures the relatedness between the head entity and the tail one, which is helpful to alleviate the overlapping of entities. Experimental results on two benchmark datasets and a self-developed dataset constructed from online military news show that the proposed MGLT achieves state-of-the-art (SOTA) performance. Compared with other typical baselines, MGLT achieves better area under curve (AUC) and [Formula: see text]-score by up to 10.46% and 6.90%, respectively. We further demonstrate the effectiveness of using ensemble learning to fully exploit complementary information from multiple MGLT-based base learners to improve the overall performance for imbalanced data classification on the military dataset. Such results indicate that the proposed ensemble learning model is effective and robust to be applied in practical applications.

The determination of standard values of vibration transmissibility at the wrist of the human body is very important to understanding vibration exposure and developing a valid model. In this work, the effects of human weight, gender, and other factors on the transmissibility of hand–arm vibration and vibration transmissibility in the [Formula: see text]-axis direction at the wrist site were randomly collected from 200 subjects (154 males and 46 females) as per the requirement of the standard ISO 10819. Moreover, the envelopes of the maximum and minimum values, as well as the mean and standard deviation of the vibration transmissibility at the wrist site, were obtained for both genders. Second, the effect of gender differences on the mean value of vibration transmissibility was analyzed. Finally, two three-degree-of-freedom (3-DOF) models of the hand–arm system were developed based on the average values of vibration transmissibility at the wrists of male and female subjects, respectively. The study revealed that the mean value of vibration transmissibility was greater in males than in females in the frequency range of 14–39 Hz. For the rest of the frequency ranges of females, their mean vibration transmissibility is greater than that of males, especially in the range 17–30 Hz. It was also observed that the vibration transmissibility of females is 10–18.5% smaller than that of males. The vibration transmissibility of females is greater than that of males by 40–83% when the frequencies are between 150 Hz and 250 Hz. The novelty of this proposed model is the prediction of model differences between males and females compared to other models that do so without considering gender differences. It is validated that the proposed modified model can be used to predict the vibration transmissibility of the wrist.

The rise in urbanization has led to an increase in high-density communities. Our study drew inspiration from the fishbone structure and applied a bionic architectural layout to optimize the wind environment within the pedestrian areas of these communities. Additionally, our study delved into the intrinsic principles and simulation-based optimization methods for the wind environment in the pedestrian areas of the fishbone bionic architectural complex. Using Computational Fluid Dynamics (CFD) simulation technology, our study simulated the building models both before and after the community renovation. Furthermore, our study simulated nine architectural complex models proposed by our study, considering the architectural layout and facade height within the community. By verifying and comparing the simulation results, our study observed that the fishbone bionic architectural form significantly improved the wind environment within the community’s pedestrian areas. Regarding the facade height, our study discovered that taller architectural units should be positioned away from the prevailing wind direction to enhance the airflow within the architectural complex. This research carries substantial implications for the renovation and planning of residential communities, particularly in large urban settings with a substantial population.

Sentiment Analysis (SA) is a computational study that examines people’s opinions, attitudes, and opinions based on their written text. Keralites’ mother tongue, Malayalam, is the most often used language to express themselves on Twitter. As there is no automatic Sentiment Analyzer in Malayalam, the SA of Twitter messages is necessary. In this research work, a Malayalam sentiment analysis model is introduced. The input raw data in the form of reviews are fed for the pre-processing stage. The pre-processing module includes sentence tokenization via Sandi Splitting and parts of speech (POS) tagging. Subsequently, Bag of Word, the Proposed weightage-based Term Frequency-Inverse Document Frequency, Unigram with the dictionary and Unigram with dictionary including negation words have been considered for feature vector formation. Finally, the review classification is undergone via the proposed hybrid classifier, which is constructed by hybridizing the Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN), respectively. To enhance the classification performance, the weight of the CNN classifier will be fine-tuned via an Improved Bumble bee mating Optimization (IBBMO), which is an advanced version of standard BBMO. The performance of the proposed work is compared over other conventional models in terms of positive, negative, convergence, and other measures. The accuracy of the proposed method is 95.436, which is much better than the existing works like DBN[Formula: see text]86.25, NN[Formula: see text]87.080, SVM[Formula: see text]87.234, WOA+Hybrid Classifier[Formula: see text]88.440, LA+Hybrid Classifier[Formula: see text]89.35285, CSO+Hybrid Classifier[Formula: see text]92.021, BBMO+Hybrid Classifier[Formula: see text]93.4248, and ML techniques[Formula: see text]94.014, at the 90th training percentage.

This work presents a novel multi-class detection approach for predicting car damage due to accidents or other reasons. The proposed prediction model intends to extract the class information to analyze the objects over the frames. Here, a novel masking-based deep convolutional neural network (MD-CNN) is proposed to capture the car parts’ regions and the classification process. The proposed model works well compared to the various existing approaches. The computation is initiated with the acquisition of damaged images, and pre-processing is performed for data labeling, which is further divided into training and testing samples. The features are internally analyzed by the proposed MD-CNN model, where the internal feature extraction process is a significant advantage with the classification process. The simulation is done in MATLAB 2020a environment where various performance metrics like accuracy, precision, [Formula: see text]-measure, recall and detection rate are evaluated and compared with conventional learning approaches. The model gives an average precision of 94.5% and [Formula: see text]1-score of 91% which is higher than other approaches. The proposed model establishes a better trade-off among the prevailing approaches with superior prediction accuracy.

Aiming at the navigation congestion of water conservancy projects, the shipowner’s psychological mechanism and behavior evolution are embedded into the dynamic model, which explores the evolution law using empirical data. The structural equation model (SEM) was constructed based on four variables: Shipowner personality, waiting time perception, congestion charging experience and anchorage environmental facilities (AEF). Taking the Three Gorges Project as an example, the influence relationship and degree are empirically tested. Then the dynamic evolution process was simulated by using the system dynamics (SD). The results suggest that the waiting time perception and congestion charge can weaken the shipowner’s intention, while the anchorage environment facilities can enhance the intention. The different personality characteristics will bring different effects, and the waiting time perception is the key intermediary. The psychological cost determines the dynamic evolution. Adjusting the factor can help the ship transfer between the navigable building and turning over the dam, realizing the balanced “double-channel” and alleviating navigation congestion.

Aiming at the shortcomings of single ant colony optimization such as many redundant nodes, slow convergence and low efficiency, based on the idea of “selection-crossover” of genetic algorithm, an improved fusion algorithm of ant colony optimization and genetic algorithm is proposed. In this paper, the fusion algorithm includes “optimal strategy” and “genetic region strategy”. The optimal strategy is that high-quality parents are selected by roulette in the first [Formula: see text] paths of each generation; genetic region strategy is that according to the path information of the parents, the grid map is divided into genetic area and nongenetic area. Genetic area refers to the area where the offspring ants can pass, and nongenetic area refers to the area where the offspring ants can’t pass; finally, the offspring ant searches the path in the genetic region to reduce the search range of the offspring ant and improve the convergence speed. Simulation results show that the fusion algorithm has faster searching speed and more stable convergence than the basic ant colony optimization and other improved ant colony optimization.

Manufacturing enterprise management mode from the traditional manufacturing production gradually transformed into service-oriented manufacturing in cloud manufacturing. As a carrier for manufacturing resources and manufacturing capabilities, the manufacturer directly affects the advantages and disadvantages of cloud manufacturing services. Therefore, manufacturer optimization has received increasing attention. This paper studies the problem of manufacturer optimization in cloud manufacturing environment, analyzes the 11 kinds of factors which affect the service quality of the manufacturer, meanwhile, it establishes the evaluation indexes system and objective function, optimization algorithm for solving the objective function, and then, it proposes optimization algorithm for solving the objective function, realizes the optimal selection of manufacturers under different demands. The simulation results show that the established evaluation indexes system can display the manufacturing resources and capabilities more fully. Based on the evaluation indexes system model, all the factors that influence the optimization of the manufacturer can be combined together, and the complex optimization problem can be transformed into the mathematical problem. The algorithm can be used to select the best manufacturer under different requirements, and to provide the best service for users.

Predicting engineering students’ academic performance is crucial for teachers to implement teaching interventions. How to use students’ behavior records to predict grades during school has attracted extensive attention from researchers. Various machine learning methods have been proposed by previous researchers to predict engineering students’ academic performance through their learning records. However, one of the major challenges now is that many researchers ignore the problem of reduced accuracy of prediction models due to imbalanced data. Simultaneously, a single classifier is susceptible to data changes and has an unstable classification performance. To this end, this paper proposes the RBF-Naive Bayes (RBFNB) prediction model to predict students’ academic performance, which integrates Radial Basis Function (RBF) Network algorithm and Naive Bayes algorithm. The method of RBF Network is applied to predict the academic performance of students and generate prediction results. Then, the method of Naive Bayes is adopted to correct the previous prediction results for improving the accuracy of the prediction. The proposed RBFNB prediction model was applied to the data set about the engineering students’ academic performance from a university and the Kalboard 360 public data set, respectively.