Electronics

Research

16 pages, 1944 KiB

Open AccessArticle

A Novel Framework for Risk Warning That Utilizes an Improved Generative Adversarial Network and Categorical Boosting

by Yan Peng, Yue Liu, Jie Wang and Xiao Li

Electronics 2024, 13(8), 1538; https://doi.org/10.3390/electronics13081538 - 18 Apr 2024

Viewed by 315

To address the problems of inadequate training and low precision in prediction models with small-sample-size and incomplete data, a novel SALGAN-CatBoost-SSAGA framework is proposed in this paper. We utilize the standard K-nearest neighbor algorithm to interpolate missing values in incomplete data, and employ [...] Read more.

To address the problems of inadequate training and low precision in prediction models with small-sample-size and incomplete data, a novel SALGAN-CatBoost-SSAGA framework is proposed in this paper. We utilize the standard K-nearest neighbor algorithm to interpolate missing values in incomplete data, and employ EllipticEnvelope to identify outliers. SALGAN, a generative adversarial network with a self-attention mechanism of label awareness, is utilized to generate virtual samples and increase the diversity of the training data for model training. To avoid local optima and improve the accuracy and stability of the standard CatBoost prediction model, an improved Sparrow Search Algorithm (SSA)–Genetic Algorithm (GA) combination is adopted to construct an effective CatBoost-SSAGA model for risk warning, in which the SSAGA is used for the global parameter optimization of CatBoost. A UCI heart disease dataset is used for heart disease risk prediction. The experimental results show the superiority of the proposed model in terms of accuracy, precision, recall, and F1-values, as well as the AUC. Full article

(This article belongs to the Special Issue Advances in Intelligent Data Analysis and Its Applications, Volume II)

► Show Figures

Figure 1

17 pages, 4069 KiB

Open AccessArticle

A Lightweight 6D Pose Estimation Network Based on Improved Atrous Spatial Pyramid Pooling

by Fupan Wang, Xiaohang Tang, Yadong Wu, Yinfan Wang, Huarong Chen, Guijuan Wang and Jing Liao

Electronics 2024, 13(7), 1321; https://doi.org/10.3390/electronics13071321 - 01 Apr 2024

Viewed by 546

Abstract

It is difficult for lightweight neural networks to produce accurate 6DoF pose estimation effects due to their accuracy being affected by scale changes. To solve this problem, we propose a method with good performance and robustness based on previous research. The enhanced PVNet-based [...] Read more.

It is difficult for lightweight neural networks to produce accurate 6DoF pose estimation effects due to their accuracy being affected by scale changes. To solve this problem, we propose a method with good performance and robustness based on previous research. The enhanced PVNet-based method uses depth-wise convolution to build a lightweight network. In addition, coordinate attention and atrous spatial pyramid pooling are used to ensure accuracy and robustness. This method effectively reduces the network size and computational complexity and is a lightweight 6DoF pose estimation method based on monocular RGB images. Experiments on public datasets and self-built datasets show that the average ADD(-S) estimation accuracy and 2D projection index of the improved method are improved. For datasets with large changes in object scale, the estimation accuracy of the average ADD(-S) is greatly improved. Full article

(This article belongs to the Special Issue Advances in Intelligent Data Analysis and Its Applications, Volume II)

► Show Figures

Figure 1

13 pages, 2044 KiB

Open AccessArticle

Deep Neural Network Confidence Calibration from Stochastic Weight Averaging

by Zongjing Cao, Yan Li, Dong-Ho Kim and Byeong-Seok Shin

Electronics 2024, 13(3), 503; https://doi.org/10.3390/electronics13030503 - 25 Jan 2024

Viewed by 857

Abstract

Overconfidence in deep neural networks (DNN) reduces the model’s generalization performance and increases its risk. The deep ensemble method improves model robustness and generalization of the model by combining prediction results from multiple DNNs. However, training multiple DNNs for model averaging is a [...] Read more.

Overconfidence in deep neural networks (DNN) reduces the model’s generalization performance and increases its risk. The deep ensemble method improves model robustness and generalization of the model by combining prediction results from multiple DNNs. However, training multiple DNNs for model averaging is a time-consuming and resource-intensive process. Moreover, combining multiple base learners (also called inducers) is hard to master, and any wrong choice may result in lower prediction accuracy than from a single inducer. We propose an approximation method for deep ensembles that can obtain ensembles of multiple DNNs without any additional costs. Specifically, multiple local optimal parameters generated during the training phase are sampled and saved by using an intelligent strategy. We use cycle learning rates starting at 75% of the training process and save the weights associated with the minimum learning rate in every iteration. Saved sets of the multiple model parameters are used as weights for a new model to perform forward propagation during the testing phase. Experiments on benchmarks of two different modalities, static images and dynamic videos, show that our method not only reduces the calibration error of the model but also improves the accuracy of the model. Full article

(This article belongs to the Special Issue Advances in Intelligent Data Analysis and Its Applications, Volume II)

► Show Figures

Figure 1

17 pages, 3838 KiB

Open AccessArticle

Visual Analysis Method for Traffic Trajectory with Dynamic Topic Movement Patterns Based on the Improved Markov Decision Process

by Huarong Chen, Yadong Wu, Huaquan Tang, Jing Lei, Guijuan Wang, Weixin Zhao, Jing Liao, Fupan Wang and Zhong Wang

Electronics 2024, 13(3), 467; https://doi.org/10.3390/electronics13030467 - 23 Jan 2024

Viewed by 624

Abstract

The visual analysis of trajectory topics is helpful for mining potential trajectory patterns, but the traditional visual analysis method ignores the evolution of the temporal coherence of the topic. In this paper, a novel visual analysis method for dynamic topic analysis of traffic [...] Read more.

The visual analysis of trajectory topics is helpful for mining potential trajectory patterns, but the traditional visual analysis method ignores the evolution of the temporal coherence of the topic. In this paper, a novel visual analysis method for dynamic topic analysis of traffic trajectory is proposed, which is used to explore and analyze the traffic trajectory topic and evolution. Firstly, the spatial information is integrated into trajectory words, calculating the dynamic trajectory topic model based on dynamic analysis modeling and, consequently, correlating the evolution of the trajectory topic between adjacent time slices. Secondly, in the trajectory topic, a representative trajectory sequence is generated to overcome the problem of the trajectory topic model not considering the word order, based on the improved Markov Decision Process. Subsequently, a set of meaningful visual codes is designed to analyze the trajectory topic and its evolution through the parallel window visual model from a spatial-temporal perspective. Finally, a case evaluation shows that the proposed method is effective in analyzing potential trajectory movement patterns. Full article

(This article belongs to the Special Issue Advances in Intelligent Data Analysis and Its Applications, Volume II)

► Show Figures

Figure 1

13 pages, 670 KiB

Open AccessArticle

RoCS: Knowledge Graph Embedding Based on Joint Cosine Similarity

by Lifeng Wang, Juan Luo, Shiqiao Deng and Xiuyuan Guo

Electronics 2024, 13(1), 147; https://doi.org/10.3390/electronics13010147 - 28 Dec 2023

Cited by 1 | Viewed by 832

Abstract

Knowledge graphs usually have many missing links, and predicting the relationships between entities has become a hot research topic in recent years. Knowledge graph embedding research maps entities and relations to a low-dimensional continuous space representation to predict links between entities. The present [...] Read more.

Knowledge graphs usually have many missing links, and predicting the relationships between entities has become a hot research topic in recent years. Knowledge graph embedding research maps entities and relations to a low-dimensional continuous space representation to predict links between entities. The present research shows that the key to the knowledge graph embedding approach is the design of scoring functions. According to the scoring function, knowledge graph embedding methods can be classified into dot product models and distance models. We find that the triple scores obtained using the dot product model or the distance model were unbounded, which leads to large variance. In this paper, we propose RotatE Cosine Similarity (RoCS), a method to compute the joint cosine similarity of complex vectors as a scoring function to make the triple scores bounded. Our approach combines the rotational properties of the complex vector embedding model RotatE to model complex relational patterns. The experimental results demonstrate that the newly introduced RoCS yields substantial enhancements compared to RotatE across various knowledge graph benchmarks, improving up to 4.0% in hits at 1 (Hits@1) on WN18RR and improving up to 3.3% in Hits@1 on FB15K-237. Meanwhile, our method achieves some new state-of-the-art (SOTA), including Hits@3 of 95.6%, Hits@10 of 96.4% on WN18, and mean reciprocal rank (MRR) of 48.9% and Hits@1 of 44.5% on WN18RR. Full article

(This article belongs to the Special Issue Advances in Intelligent Data Analysis and Its Applications, Volume II)

► Show Figures

Figure 1

27 pages, 8733 KiB

Open AccessArticle

Improved A-Star Path Planning Algorithm in Obstacle Avoidance for the Fixed-Wing Aircraft

by Jing Li, Chaopeng Yu, Ze Zhang, Zimao Sheng, Zhongping Yan, Xiaodong Wu, Wei Zhou, Yang Xie and Jun Huang

Electronics 2023, 12(24), 5047; https://doi.org/10.3390/electronics12245047 - 18 Dec 2023

Viewed by 660

Abstract

The flight management system is a basic component of avionics for modern airliners. However, the airborne flight management system needs to be improved and relies on imports; path planning is the key to the flight management system. Based on the classical A* algorithm, [...] Read more.

The flight management system is a basic component of avionics for modern airliners. However, the airborne flight management system needs to be improved and relies on imports; path planning is the key to the flight management system. Based on the classical A* algorithm, this paper proposes an improved A* path planning algorithm, which solves the problem of low planning efficiency and following a non-smooth path. In order to solve the problem of the large amount of data calculation and long planning time of the classical A* algorithm, a new data structure called a “value table” is designed to replace the open table and close table of the classical A* algorithm to improve the retrieval efficiency, and the Heap sort algorithm is used to optimize the efficiency of node sorting. Aiming at the problem that the flight trajectory is hard to follow, the trajectory smoothing optimization algorithm combined with turning angle limit is proposed. The gray value in the digital map is added to the A* algorithm, and the calculation methods of gray cost, cumulative cost, and estimated cost are improved, which can better meet the constraints of obstacle avoidance. Through the comparative simulation verification of the algorithm, the improved A* algorithm can significantly reduce the path planning time to 1% compared to the classical A* algorithm; it can be seen that the proposed algorithm improves the efficiency of path planning and the smoother planned path, which has obvious advantages compared to the classical A* algorithm. Full article

(This article belongs to the Special Issue Advances in Intelligent Data Analysis and Its Applications, Volume II)

► Show Figures

Figure 1

14 pages, 2551 KiB

Open AccessArticle

Joint Overlapping Event Extraction Model via Role Pre-Judgment with Trigger and Context Embeddings

by Qian Chen, Kehan Yang, Xin Guo, Suge Wang, Jian Liao and Jianxing Zheng

Electronics 2023, 12(22), 4688; https://doi.org/10.3390/electronics12224688 - 18 Nov 2023

Viewed by 789

Abstract

The objective of event extraction is to recognize event triggers and event categories within unstructured text and produce structured event arguments. However, there is a common phenomenon of triggers and arguments of different event types in a sentence that may be the same [...] Read more.

The objective of event extraction is to recognize event triggers and event categories within unstructured text and produce structured event arguments. However, there is a common phenomenon of triggers and arguments of different event types in a sentence that may be the same word elements, which poses new challenges to this task. In this article, a joint learning framework for overlapping event extraction (ROPEE) is proposed. In this framework, a role pre-judgment module is devised prior to argument extraction. It conducts role pre-judgment by leveraging the correlation between event types and roles, as well as trigger embeddings. Experiments on the FewFC show that the proposed model outperforms other baseline models in terms of Trigger Classification, Argument Identification, and Argument Classification by 0.4%, 0.9%, and 0.6%. In scenarios of trigger overlap and argument overlap, the proposed model outperforms the baseline models in terms of Argument Identification and Argument Classification by 0.9%, 1.2%, 0.7%, and 0.6%, respectively, indicating the effectiveness of ROPEE in solving overlapping events. Full article

(This article belongs to the Special Issue Advances in Intelligent Data Analysis and Its Applications, Volume II)

► Show Figures

Figure 1

23 pages, 1650 KiB

Open AccessArticle

Resolving Agent Conflicts Using Enhanced Uncertainty Modeling Tools for Intelligent Decision Making

by Yanhui Zhai, Zihan Jia and Deyu Li

Electronics 2023, 12(21), 4547; https://doi.org/10.3390/electronics12214547 - 05 Nov 2023

Viewed by 801

Abstract

Conflict analysis in intelligent decision making has received increasing attention in recent years. However, few researchers have analyzed conflicts by considering trustworthiness from the perspective of common agreement and common opposition. Since L-fuzzy three-way concept lattice is able to describe both the [...] Read more.

Conflict analysis in intelligent decision making has received increasing attention in recent years. However, few researchers have analyzed conflicts by considering trustworthiness from the perspective of common agreement and common opposition. Since L-fuzzy three-way concept lattice is able to describe both the attributes that objects commonly possess and the attributes that objects commonly do not possess, this paper introduces an L-fuzzy three-way concept lattice to capture the issues on which agents commonly agree and the issues which they commonly oppose, and proposes a hybrid conflict analysis model. In order to resolve conflicts identified by the proposed model, we formulate the problem as a knapsack problem and propose a method for selecting the optimal attitude change strategy. This strategy takes into account the associated costs and aims to provide the decision maker with the most favorable decision in terms of resolving conflicts and reaching consensus. To validate the effectiveness and feasibility of the proposed model, a case study is conducted, providing evidence of the model’s efficacy and viability in resolving conflicts. Full article

(This article belongs to the Special Issue Advances in Intelligent Data Analysis and Its Applications, Volume II)

► Show Figures