Search

Multi-category text classification aims to assign labels under multiple categories to each sample in the dataset. However, most researchers overlook the relationship between labels and do not effectively incorporate text and label information. To address these problems, we propose a label embedding joint cross-attention (LEJCA) model for multi-category text classification using label relations and text information. Specifically, we first input the characters of the text into the Pre-training Language Model to obtain the representation of the text. In addition, a label embedding method based on graph representation is proposed to realize multi-dimensional mining and quantitative description of label relations. Finally, a joint cross-attention (JCA) model is proposed to integrate text and label embedding representations. Experimental results on our own constructed dataset E-government document show that LEJCA ’s micro-average F1 score is improved by $9.7~{\%}$ compared to the baseline’s best result.

The Antagonistic Weapon-Target Assignment (AGWTA) problem is a crucial decision issue in Command & Control (C2). Since this is a minimax problem, co-evolutionary algorithms can be used to solve it effectively. However, the co-evolutionary algorithm is originally designed for continuous minimax problems which loses its efficiency to discrete contexts. In this paper, a novel elitism co-evolutionary algorithm is proposed to solve the AGWTA. Firstly, an improved AGWTA model for air combat based on the attack and evasion strategies is proposed. Secondly, an elite cooperative genetic algorithm based on the framework of the co-evolutionary algorithm is put forward. In this proposed algorithm, a problem-specific coding method and evolution operator are designed. Meanwhile, an elite individual update mechanism is presented. Finally, based on the analysis of the relationship between the feasible solutions under the air combat environment, an evaluation index is proposed. Experiments show that the proposed algorithm has higher accuracy than traditional co-evolutionary algorithms for solving AGWTA problems.

To compare CNN models implemented using different strategies in the CT assessment of EGFR mutation status in patients with lung adenocarcinoma. 1,010 consecutive lung adenocarcinoma patients with known EGFR mutation status were randomly divided into a training set (n = 810) and a testing set (n = 200). The CNN models were constructed based on ResNet-101 architecture but implemented using different strategies: dimension filters (2D/3D), input sizes (small/middle/large and their fusion), slicing methods (transverse plane only and arbitrary multi-view planes), and training approaches (from scratch and fine-tuning a pre-trained CNN). The performance of the CNN models was compared using AUC. The fusion approach yielded consistently better performance than other input sizes, although the effect often did not reach statistical significance. Multi-view slicing was significantly superior to the transverse method when fine-tuning a pre-trained 2D CNN but not a CNN trained from scratch. The 3D CNN was significantly better than the 2D transverse plane method but only marginally better than the multi-view slicing method when trained from scratch. The highest performance (AUC = 0.838) was achieved for the fine-tuned 2D CNN model when built using the fusion input size and multi-view slicing method. The assessment of EGFR mutation status in patients is more accurate when CNN models use more spatial information and are fine-tuned by transfer learning. Our finding of the implementation strategy of a CNN model could be a guide to other medical 3D images applications. Compared with other published studies which used medical images to identify EGFR mutation status, our CNN model achieved the best performance in the biggest patient cohort.

Degradation modeling and remaining useful life (RUL) prediction for products with multiple degradation features are hot topics in the prognostic and health management. The key to this problem is to describe the dependence among multiple degradation features effectively. In this paper, a multivariate degradation modeling approach based on the Bayesian dynamic linear model (BDLM) is proposed to calculate the RULs of degradation features, and the Copula function is employed to capture the dependence among RUL distributions. A combined BDLM is used to establish the multivariate degradation model, which includes two typical BDLMs, namely, the linear growth model and seasonal factors model. After the model parameters get calibrated by the maximum likelihood estimation, the model can predict the degradation process of features. Once the failure thresholds are given, the probability density function and cumulative distribution function (CDF) of RUL for each degradation feature can be obtained. Since these RUL distributions are not independent of each other, the Copula function is adopted herein to couple the CDFs. Finally, some practical testing data of a microwave component, which has two degradation features, are utilized to validate our proposed method. This paper provides a new idea for the multivariate degradation modeling and RUL prediction.

Accelerated degradation testing (ADT) is commonly used to obtain degradation data of products by exerting loads over usage conditions. Such data can be used for estimating component lifetime and reliability under usage conditions. The design of ADT entails to establish a model of the degradation process and define the test plan to satisfy given criteria under the constraint of limited test resources. Bayesian optimal design is a method of decision theory under uncertainty, which uses historical data and expert information to find the optimal test plan. Different expected utility functions can be selected as objectives. This paper presents a method for Bayesian optimal design of ADT, based on the inverse Gaussian process and considering three objectives for the optimization: relative entropy, quadratic loss function, and Bayesian D-optimality. The Markov chain Monte Carlo and the surface fitting methods are used to obtain the optimal plan. By sensitivity analysis and a proposed efficiency factor, the Bayesian D-optimality is identified as the most robust and appropriate objective for Bayesian optimization of ADT.

Recently, Generative Adversarial Network (GAN) based approaches are applied in facial attribute translation. However, many tasks, i.e. multi facial attributes translation and background invariance, are not well handled in the literature. In this paper, we propose a novel GAN-based method that aims to get the target image that performs better within modifying one or more facial attributes in a single model. The model generator learns multi-points by inputting a re-coded transfer vector, ensuring the single model could learn multiple attributes simultaneously. It also optimizes the cycle loss to enhance the efficiency of transferring multi attributes. Moreover, the method uses the adaptive parameter to improve the calculation method of the loss function of the residual image. The results are also compared with the StarGAN v2, which is the current state-of-the-art model to prove the effectiveness and advancedness. Experiments show that our method has a satisfactory performance in multi facial attributes translation.

The key to achieving high-quality and practical van der Waals heterostructure devices made from various two-dimensional (2D) materials lies in the efficient control over clean and flexible interfaces. Inspired by the “movable-type printing”, one of the four great inventions of ancient China, we demonstrate the “movable-type” transfer and stacking of 2D materials, which utilizes prefabricated polyvinyl alcohol (PVA) film to engineer the interfacial adhesion to 2D materials, and provides a flexible, efficient and batchable transfer scheme for 2D materials. The experiments also verify the “movable-type” transfer can preciously control the position and orientation of 2D materials, which meets the burgeoning requirements such as the preparation of twisted graphene and other heterostructures. Importantly, water-solubility of PVA film ensures an ideal interface of the materials without introducing contamination. We illustrate the superiority of this method with a WSe2 vertical spin valve device, whose performance verifies the applicability and advantages of such a method for spintronics. Our PVA-assisted “movable-type” transfer process may promote the development of high-performance 2D-material-based devices.

Automatic human gender recognition is an important and classical problem in artificial intelligence. Most of the previous gender recognition works are based on vision appearance and biometric characteristics. However, there are fewer gender recognition approaches for 3D human shapes. In this article, we propose a novel deep neural network learning method for gender recognition of 3D human shapes. Firstly, we introduce effective descriptors to distinguish male and female of 3D human shapes via probability distributions of biharmonic distances among points. Secondly, the above distances-based low-level descriptors are fed into a fully connected neural network for gender recognition. Furthermore, we construct a larger 3D human shape dataset for evaluation of the proposed gender recognition method by collecting and labeling human shape models. Compared with previous works, our method obtains higher recognition accuracy and has more advantages, such as posture invariant, robust to noises, and no need of landmarks or pre-alignment process.

The cucumber fruits have the same color with leaves and their shapes are all long and narrow, which is different from other common fruits, such as apples, tomatoes, and strawberries, etc. Therefore, cucumber fruits are more difficult to be detected by machine vision in greenhouses for special color and shape. A pixel-wise instance segmentation method, mask region-based convolutional neural network (Mask RCNN) of an improved version, is proposed to detect cucumber fruits. Resnet-101 is selected as the backbone of Mask RCNN with feature pyramid network (FPN). To improve the detection precision, region proposal network (RPN) in original Mask RCNN is improved. Logical green (LG) operator is designed to filter non-green background and limit the range of anchor boxes. Besides, the scales and aspect ratios of anchor boxes are also adjusted to fit the size and shape of fruits. Improved Mask RCNN has a better performance on test images. The test results are compared with that of original Mask RCNN, Faster RCNN, you only look once (YOLO) V2 and YOLO V3. The F1 score of improved Mask RCNN in test results reaches 89.47%, which is higher than the other methods. The average elapsed time of improved Mask RCNN is 0.3461 s, which is only lower than the original Mask RCNN. Meanwhile, the mean value and standard deviation of location deviation in improved Mask RCNN are 2.10 pixels and 1.73 pixels respectively, which are lower than the other methods.

The skins of most mature apple fruits are incompletely red and also include green and pale yellow color, which increases the difficulty of fruit detection by machine vision. A detection method based on color and shape features is proposed for this kind of apple fruits. Simple linear iterative clustering (SLIC) is adapted to segment images taken in orchards into super-pixel blocks. The color feature extracted from blocks is used to determine candidate regions, which can filter a large proportion of non-fruit blocks and improve detection precision. Next, the histogram of oriented gradient (HOG) is adopted to describe the shape of fruits, which is applied to detect fruits in candidate regions and locate the position of fruits further. The proposed method was tested by images taken under different illuminations. The average values of recall, precision, and F1 reach 89.80%, 95.12%, and 92.38% respectively. The performance of detecting fruits covered at different levels is also tested. The values of the recall are all more than 85%, which indicates that proposed method can detect a great part of covered fruits. Compared with pedestrian detection method and faster region-based convolutional neural network (RCNN), the proposed method has the best performance and higher than faster RCNN slightly. However, the proposed method is not robust to noise and its elapsed time of one image is 1.94 s and less than faster RCNN.

Public emergency occurs every day, and the information diffusion of an online social network (OSN) shows the characteristics of large capacity, a wide range of far-reaching effects. In order to analyze the role of opinion leader in OSN and the life cycle of Sina Microblog information diffusion, the public emergency information diffusion model and a new method grading opinion leader model are proposed. Firstly, the people of OSN are divided into three kinds: supporter, neutral, and objector. Information interaction formula of opinion leader is built up according to the properties of all kinds of subjects (public people, rumor maker, and opinion leader), and the statistical analysis of the microblog information data is obtained by the computer technology. Second, the communication power is constructed as a function related to the three factors (the number of forwarding, the degree of activity, and the number of fans), and then a weight calculation method based on the analytic hierarchy process is established. Communication power is combined with the new proposed model, and the information diffusion of opinion leader and microblog life cycle is researched. The simulation results show that the established public emergency information diffusion model can truly reflect the actual situation. Opinion leaders have a strong communication in microblog information diffusion. Opinion leaders can influence those who are unsteady about certain public emergency. The value of error is 13.7% and 5.1% by comparing the true data. It is proved that the new proposed models are reasonable and effective for the communication life of the analysis opinion leader in microblog information diffusion.