Two-stage Framework Research Articles

A two-stage game-theoretical framework for Pricing & Management problem of shared mobility in a competitive market

ABSTRACT With the global proliferation of shared mobility systems, the vehicle-sharing market has witnessed an influx of companies and investors, thereby intensifying competition and potentially leading to suboptimal decision-making. Consequently, there is a pressing need for a modelling framework that specifically addresses operational and strategic-level policy design for multiple vehicle-sharing companies in a competitive market. In this study, we develop a two-stage game-theoretical framework to simulate the competitive process between multiple vehicle-sharing companies, facilitating their business decisions regarding pricing and management (P&M) problem, where management mainly include service area design, vehicle deployment, redirecting, and rebalancing. In the first stage of the framework, two companies operating under the same shared mobility service form a strategic alliance to compete against another vehicle-sharing service. Subsequently, in the second stage, these two companies engage in competition to maximise their individual profits. Based on the factors such as generalised travel cost, average acceptance level and vehicle availability, the competition process of each stage is formulated as a generalised Nash equilibrium problem. A linearisation method is employed for solving the equilibrium solution. Finally, we apply our proposed models to various vehicle-sharing systems and application scenarios to demonstrate their applicability. The numerical results illustrate that our model assists companies in making rational management decisions amidst intense competition while also revealing market competition landscape of vehicle-sharing systems based on different travel distances in the shared mobility market.

Groundwater Contamination Source Recognition Based on a Two-Stage Inversion Framework with a Deep Learning Surrogate

Groundwater contamination source recognition is an important prerequisite for subsequent remediation efforts. To overcome the limitations of single inversion methods, this study proposed a two-stage inversion framework by integrating two primary inversion approaches—simulation-optimization and simulation-data assimilation—thereby enhancing inversion accuracy. In the first stage, the ensemble smoother with multiple data assimilation method (a type of simulation-data assimilation) conducted a global broad search to provide better initial values and ranges for the second stage. In the subsequent stage, a collective decision optimization algorithm (a type of simulation-optimization) was used for a refined deep search, further enhancing the final inversion accuracy. Additionally, a deep learning method, the multilayer perceptron, was utilized to establish a surrogate of the simulation model, reducing computational costs. These theories and methods were applied and validated in a hypothetical scenario for the synchronous identification of the contamination source and boundary conditions. The results demonstrated that the proposed two-stage inversion framework significantly improved search accuracy compared to single inversion methods, with a mean relative error and mean absolute error of just 4.95% and 0.1756, respectively. Moreover, the multilayer perceptron surrogate model offered greater approximation accuracy to the simulation model than the traditional shallow learning surrogate model. Specifically, the coefficient of determination, mean relative error, mean absolute error, and root mean square error were 0.9860, 9.72%, 0.1727, and 0.47, respectively, highlighting its significant advantages. The findings of this study can provide more reliable technical support for practical case applications and improve subsequent remediation efficiency.

OccTr: A Two-Stage BEV Fusion Network for Temporal Object Detection

Temporal fusion approaches are critical for 3D visual perception tasks in IOV (Internet of Vehicles), but they often rely on intermediate representations without fully utilizing position information from the previous frame’s detection results, which cannot compensate for the lack of depth information in visual data. In this work, we propose a novel framework called OccTr (Occupancy Transformer) that combines two temporal cues, intermediate representation and back-end representation, via occupancy map to enhance temporal fusion in object detection task. OccTr leverages attention mechanisms to perform both intermediate and back-end temporal fusion by incorporating intermediate BEV (bird’s-eye view) features and back-end prediction results of the detector. Our two-stage framework includes occupancy map generation and cross-attention feature fusion. In stage one, the prediction results are converted into occupancy grid map format to generate back-end representation. In stage two, the high-resolution occupancy maps are fused with BEV features using cross-attention layers. This fused temporal cue provides a strong prior for the temporal detection process. Experimental results demonstrate the effectiveness of our method in improving detection performance, achieving an NDS (nuScenes Detection Score) metric score of 37.35% on the nuScenes test set, which is 1.94 points higher than the baseline.

Cropland inundation mapping in a mountain dominated region based on multi-resolution remotely sensed imagery and active learning for semantic segmentation

Accurate and timely mapping of cropland inundated area is critical to loss assessment and recovery plan development. However, existing methods for cropland inundation mapping using Synthetic Aperture Radar (SAR) imagery are primarily designed for plain regions and may not be suitable for mountainous regions. The complex rugged terrain at large scale present challenges such as significant noise from radar shadows, smooth surfaces, floodplains, and permanent water bodies. To remove these noise for effectively mapping of cropland inundation, this study proposes a robust two-stage framework integrating the multi-resolution remotely sensed imageries and semantic segmentation method. First, active learning for cropland extraction is performed with Google Very-High-Resolution (VHR) imagery and Segformer model. Second, the cropland inundation area is obtained using marker-controlled watershed segmentation (MCW) on Sentinel-1 SAR imagery with area-of-interest (AOI) filtering. Results show that the extracted cropland mask are superior to six public cropland data products including ESA, ESRI, FCS30, FROM, GL30, and SLOPE. In addition, with AOI filtering, the F-score of the potential flood identification using MCW was increased from 0.685 to 0.828. At last, the monthly cropland inundation caused by intense rainfall varied approximately from 6.6 km2 to 436.1 km2 during monsoon season of 2017–2022 in Guangdong. Our approach offers a fast and feasible solution for mapping cropland inundation in large and complex mountainous regions, without requiring extensive parameter tuning or additional data.

CapsEnhancer: An Effective Computational Framework for Identifying Enhancers Based on Chaos Game Representation and Capsule Network.

Enhancers are a class of noncoding DNA, serving as crucial regulatory elements in governing gene expression by binding to transcription factors. The identification of enhancers holds paramount importance in the field of biology. However, traditional experimental methods for enhancer identification demand substantial human and material resources. Consequently, there is a growing interest in employing computational methods for enhancer prediction. In this study, we propose a two-stage framework based on deep learning, termed CapsEnhancer, for the identification of enhancers and their strengths. CapsEnhancer utilizes chaos game representation to encode DNA sequences into unique images and employs a capsule network to extract local and global features from sequence "images". Experimental results demonstrate that CapsEnhancer achieves state-of-the-art performance in both stages. In the first and second stages, the accuracy surpasses the previous best methods by 8 and 3.5%, reaching accuracies of 94.5 and 95%, respectively. Notably, this study represents the pioneering application of computer vision methods to enhancer identification tasks. Our work not only contributes novel insights to enhancer identification but also provides a fresh perspective for other biological sequence analysis tasks.

Multimodal Machine Learning for Prognosis and Survival Prediction in Renal Cell Carcinoma Patients: A Two-Stage Framework with Model Fusion and Interpretability Analysis

Current medical limitations in predicting cancer survival status and time necessitate advancements beyond traditional methods and physical indicators. This research introduces a novel two-stage prognostic framework for renal cell carcinoma, addressing the inadequacies of existing diagnostic approaches. In the first stage, the framework accurately predicts the survival status (alive or deceased) with metrics Accuracy, Precision, Recall, and F1 score to evaluate the effects of the classification results, while the second stage focuses on forecasting the future survival time of deceased patients with Root Mean Square Error and Mean Absolute Error to evaluate the regression results. Leveraging popular machine learning models, such as Adaptive Boosting, Extra Trees, Gradient Boosting, Random Forest, and Extreme Gradient Boosting, along with fusion models like Voting, Stacking, and Blending, our approach significantly improves prognostic accuracy as shown in our experiments. The novelty of our research lies in the integration of a logistic regression meta-model for interpreting the blending model’s predictions, enhancing transparency. By the SHapley Additive exPlanations’ interpretability, we provide insights into variable contributions, aiding understanding at both global and local levels. Through modal segmentation and multimodal fusion applied to raw data from the Surveillance, Epidemiology, and End Results program, we enhance the precision of renal cell carcinoma prognosis. Our proposed model provides an interpretable analysis of model predictions, highlighting key variables influencing classification and regression decisions in the two-stage renal cell carcinoma prognosis framework. By addressing the black-box problem inherent in machine learning, our proposed model helps healthcare practitioners with a more reliable and transparent basis for applying machine learning in cancer prognostication.

Integrated real-time signal control and routing optimization: A two-stage rolling horizon framework with decentralized solution

This paper presents an integrated framework for optimizing signal control and vehicle routing. An important feature of the proposed framework is the ability to simultaneously determine signal states and individual vehicle routes in real time. The general objective is to minimize the network travel time, which can be represented as a trade-off between the total route length of all vehicles and traffic conditions at signalized intersections. A two-stage rolling horizon framework is proposed to explicitly describe the relationship between individual vehicle routes and predicted traffic flow dynamics at signalized intersections. The first stage involves a signal optimization problem, while the second stage optimizes a joint signal control and vehicle routing problem. Both stages are formulated as mixed integer linear programming problems. The optimization procedure is decentralized, and the effects of vehicle routing on control performance is considered by incorporating the route length cost into the objective function. Simulation experiments validate the advantages of the proposed framework over advanced signal control strategies and dynamic user-optimal routing strategies in various scenarios. The effectiveness in improving network capacity, alleviating spillback, and decreasing congestion dissipation time under over-saturation conditions is discussed. The results of vehicle routing suggest that the total travel time can be reduced at a low rerouting cost. Sensitivity analyses demonstrate the network control performance under different compliance rates and model coefficients. Moreover, the computational feasibility of the framework is verified.

ParInfoGPT: An LLM-based two-stage framework for reliability assessment of rotating machine under partial information

Data-driven approaches on large-amount labelled data have been actively implemented for reliability assessment of rotating machine, i.e., with limited labelled data samples. Recently, the study of large language models (LLMs) has demonstrated outstanding performance in the field of natural language processing. Inspired by the LLM for addressing sequential data, this article investigates the implementation of LLM under partial information for reliability assessment of rotating machine. A novel LLM-based two-stage framework, called ParInfoGPT, is proposed by integrating a self-supervised reconstruction network and a weakly supervised classification network. Additionally, a mutual information (MI)-based informative masking strategy for pre-training and a parallel side-adapter (PSA) for fine-tuning are designed to effectively learn the proposed framework. Experiments are systematically conducted and evaluated on two real-world datasets of rotating machine. The experimental results of the proposed methodology demonstrate its superior performance on fault diagnosis under partial information for reliability assessment.

MocFormer: A Two-Stage Pre-training-Driven Transformer for Drug–Target Interactions Prediction

Drug–target interactions is essential for advancing pharmaceuticals. Traditional drug–target interaction studies rely on labor-intensive laboratory techniques. Still, recent advancements in computing power have elevated the importance of deep learning methods, offering faster, more precise, and cost-effective screening and prediction. Nonetheless, general deep learning methods often yield low-confidence results due to the complex nature of drugs and proteins, bias, limited labeled data, and feature extraction challenges. To address these challenges, a novel two-stage pre-trained framework is proposed for drug–target interactions prediction. In the first stage, pre-trained molecule and protein models develop a comprehensive feature representation, enhancing the framework’s ability to handle drug and protein diversity. This also reduces bias, improving prediction accuracy. In the second stage, a transformer with bilinear pooling and a fully connected layer enables predictions based on feature vectors. Comprehensive experiments were conducted using public datasets from DrugBank and Epigenetic-regulators datasets to evaluate the framework’s effectiveness. The results demonstrate that the proposed framework outperforms the state-of-the-art methods regarding accuracy, area under the receiver operating characteristic curve, recall, and area under the precision-recall curve. The code is available at: https://github.com/DHCGroup/MocFormer.

Optimal capacity and multi-stable flexible operation strategy of green ammonia systems: Adapting to fluctuations in renewable energy

The potential of power-to-ammonia is increasingly recognized as a large-scale renewable electrical energy storage technology in the energy-transition landscape. Unlike conventional, continuously operating Haber-Bosch production processes, power-to-ammonia processes face operational challenges stemming from fluctuations in renewable power supply. Systematic strategies that can adapt to these fluctuations are required to improve the operational flexibility and stability of power-to-ammonia processes for industrial applications. This study proposes a multi-stable flexible ammonia operation strategy, which is practical and feasible in engineering. To maximize net profit, a design scheduling coordination optimization model is established and solved using a two-stage algorithmic framework of particle swarm and mixed integer linear programming. Additionally, three flexible scenarios are designed for the simulation analysis of a grid-connected green ammonia system. Compared with the traditional constant inflexible scenario, the multi-stable flexible scenario increases annual net profit by 5.05 M$ and reduces carbon emission intensity by 1.61 kg CO2/kgNH3. Furthermore, compared with the continuous flexible scenario, the volatility of the ammonia production load is reduced by 79.89 %. The multi-stable flexible operation strategy balances the green ammonia system’s economic, safety, and low-carbon attributes. Sensitivity analysis reveals that wind and photovoltaic complementary power generation reduces green ammonia costs and carbon emissions. When the selling price of ammonia exceeds 550 $/t, ammonia production and sales should be prioritized, but at lower ammonia prices, selling electricity to the grid to optimize economic efficiency is advised. This study provides novel insights into the flexible operation of the power-to-ammonia process and is expected to offer guidance for constructing and operating green ammonia systems.

A two-stage framework for parking search behavior prediction through adversarial inverse reinforcement learning and transformer

Parking scenarios are spatially dense and have a lot of interactions, making predicting vehicles’ search behavior crucial and challenging for autonomous driving. Existing data-driven prediction methods struggle to determine vehicles’ intents and consider the surrounding environment accurately. This study proposes a novel two-stage framework for parking search behavior prediction, involving parking intent and vehicle trajectory predictions based on imitation learning and deep learning. First, we develop an adversarial inverse reinforcement learning model for parking search intent (PSI-AIRL) learning from measured trajectory data in an actual parking lot. Then, we design an integrated convolutional neural network (CNN) and transformer model to forecast vehicle trajectory using historical observations and the predicted parking search intents. This two-stage framework achieves parking search intent prediction by applying global information about the parking lot while improving vehicle trajectory prediction’s accuracy and robustness. Finally, the experiments are conducted on the Dragon Lake Parking (DLP) dataset to compare our framework with state-of-the-art models. The results show that our model outperforms other baseline models in accuracy for parking intent and vehicle trajectory predictions. Moreover, our model shows exceptional accuracy and robustness in predicting across diverse parking scenarios.

Lego-like composite laminate construction and analysis on the fly

High-fidelity finite element (FE) analysis is computational demanding for parametric studies and design optimizations of composite laminates because it is time-consuming and inflexible in reconfiguration. Hence, we aim to develop rapid yet accurate 3-D FE analysis that can readily change (i) layer materials including fiber angles, (ii) the number of layers, and (iii) the stacking sequence of layers. To achieve these three features, we propose an offline-online framework hinging on static condensation (SC) integrated with port reduction (PR) and reduced basis element (RBE) methods, resulting in port-reduced static condensation (PR-SC) and port-reduced static condensation reduced basis element (PR-SCRBE) analyses. Thanks to the PR-SC and PR-SCRBE techniques, we may easily build a composite laminate using a variety of layer materials and any desired number of layers, akin to assembling Lego bricks, and swiftly analyze the constructed composite laminate. For demonstration, we assess both PR-SC and PR-SCRBE approximations using three different composite laminates with varying numbers of layers for each of three-point bending and tensile tests. Compared with FE analysis, PR-SC analysis exhibits insignificant accuracy loss in displacement and stress evaluation while obtaining meaningful computational speedup. In contrast, PR-SCRBE analysis shows moderate accuracy loss in displacement and stress evaluation while gaining considerable computational speedup. In conclusion, we recommend employing PR-SC for accuracy and PR-SCRBE for efficiency as per application requirements. The proposed two-stage framework may benefit many-query and real-time applications involving composite laminate analysis, including design optimization and virtual testing.

A two-stage electricity consumption forecasting method integrated hybrid algorithms and multiple factors

Electricity consumption forecasting provides a reference basis for electricity scheduling, it has become a research hotspot in the field of electric power. The fluctuation of electricity consumption is influenced by multiple factors, which has strong randomness and are strongly disturbed by noise. Single algorithms cannot ensure the stability and accuracy of electricity consumption forecasting. To address this problem, a two-stage framework for electricity consumption forecasting is proposed in this paper. In the first stage, multi-forecasting algorithms are considered to eliminate noise and forecast the electricity consumption trend. A hybrid model is constructed, which consists of three parts: Ensemble Empirical Mode Decomposition (EEMD), Gated Recurrent Unit (GRU) and Gradient Boosting Regression Tree (GBRT). The EEMD method is employed to decompose the electricity consumption signal into different frequency components. GRU and GBRT are adopted to forecast high-frequency and low-frequency components respectively. In the second stage, multidimensional meteorological factors are considered to further improve the accuracy of the electricity consumption forecasting. A structure for feature extraction is constructed to reduce the dimension of meteorological factors. The experimental results indicate that the two-stage framework model outperforms mainstream measurement models by an average reduction of 35.7% and 43.2% in MAE and RMSE.

Li-ion battery state of health prediction through metaheuristic algorithms and genetic programming

Predicting the State of Health (SOH) of the Lithium-ion battery with higher accuracy and reduced cost is a challenging and crucial task for ensuring its reliability and safety. To achieve this, a two-stage prediction framework is proposed to find a concise expression of SOH using the health features of the battery through metaheuristic algorithms and genetic programming (GP).In Stage-I, three conflicting objectives are considered concurrently: the root-mean-square error (RMSE) of SOH prediction, the health features selected, and the expressional complexity of the battery capacity. The wrapper structure is utilized for SOH prediction, where a binary multi-objective grey wolf optimization (Binary MOGWO) algorithm is employed to select features and generate the Pareto set. Genetic programming is used to calculate the SOH value using symbolic regression models. In Stage-II, the final compromise solution is filtered from the Pareto set through decision-making approaches. The relationships between selected features and the capacity of the battery are investigated. The NASA Prognostics Center of Excellence battery dataset is chosen to verify the effectiveness of the proposed framework. The simulation results show that the features found through the framework can provide SOH predictions in the form of symbolic equations with higher accuracy, lower cost, and reduced complexity.

Efficient recognition of fish feeding behavior: A novel two-stage framework pioneering intelligent aquaculture strategies

In the domain of aquaculture, the act of feeding fish is a pivotal factor influencing both the growth of the fish and the associated cultivation costs. The implementation of intelligent feeding strategies is a crucial prerequisite for maintaining fish health and minimizing costs, with the accurate discernment of fish feeding behavior serving as the fundamental basis for the realization of such strategies. Addressing the issues of high data redundancy and substantial noise content inherent in the datasets utilized by existing identification models, as well as the intricate design and suboptimal execution efficiency of the model structures, this study introduces a two-stage framework for discriminating fish feeding behavior. In the initial stage, a re-parameterizable multi-scale object detection model is established, facilitating the acquisition of the spatial distribution of fish schools. This process effectively eliminates redundant data and noise, decouples the training and inference processes of the model, equivalently transforms the complex network training weights, and simplifies the inference process of the model. An analysis of the dataset characteristics is conducted, leading to the optimization of the model detector’s design and a reduction in both the model parameters and computational requirements. In the subsequent stage, responding to the key data characteristics of the spatial distribution of fish schools, a lightweight behavior recognition model is designed. This model enables the rapid and accurate identification of fish feeding behavior. A plethora of experimental results demonstrate that the proposed method can achieve a high recognition accuracy (Acc 83.33%) while operating under the constraints of minimal model parameters and computations (6.45M Params, 8.135G FLOPs). This provides a robust model foundation for the industrial application of the algorithm, underscoring the significant potential of the proposed method in advancing intelligent feeding strategies in aquaculture.

Data-driven crop growth simulation on time-varying generated images using multi-conditional generative adversarial networks

BackgroundImage-based crop growth modeling can substantially contribute to precision agriculture by revealing spatial crop development over time, which allows an early and location-specific estimation of relevant future plant traits, such as leaf area or biomass. A prerequisite for realistic and sharp crop image generation is the integration of multiple growth-influencing conditions in a model, such as an image of an initial growth stage, the associated growth time, and further information about the field treatment. While image-based models provide more flexibility for crop growth modeling than process-based models, there is still a significant research gap in the comprehensive integration of various growth-influencing conditions. Further exploration and investigation are needed to address this gap.MethodsWe present a two-stage framework consisting first of an image generation model and second of a growth estimation model, independently trained. The image generation model is a conditional Wasserstein generative adversarial network (CWGAN). In the generator of this model, conditional batch normalization (CBN) is used to integrate conditions of different types along with the input image. This allows the model to generate time-varying artificial images dependent on multiple influencing factors. These images are used by the second part of the framework for plant phenotyping by deriving plant-specific traits and comparing them with those of non-artificial (real) reference images. In addition, image quality is evaluated using multi-scale structural similarity (MS-SSIM), learned perceptual image patch similarity (LPIPS), and Fréchet inception distance (FID). During inference, the framework allows image generation for any combination of conditions used in training; we call this generation data-driven crop growth simulation.ResultsExperiments are performed on three datasets of different complexity. These datasets include the laboratory plant Arabidopsis thaliana (Arabidopsis) and crops grown under real field conditions, namely cauliflower (GrowliFlower) and crop mixtures consisting of faba bean and spring wheat (MixedCrop). In all cases, the framework allows realistic, sharp image generations with a slight loss of quality from short-term to long-term predictions. For MixedCrop grown under varying treatments (different cultivars, sowing densities), the results show that adding these treatment information increases the generation quality and phenotyping accuracy measured by the estimated biomass. Simulations of varying growth-influencing conditions performed with the trained framework provide valuable insights into how such factors relate to crop appearances, which is particularly useful in complex, less explored crop mixture systems. Further results show that adding process-based simulated biomass as a condition increases the accuracy of the derived phenotypic traits from the predicted images. This demonstrates the potential of our framework to serve as an interface between a data-driven and a process-based crop growth model.ConclusionThe realistic generation and simulation of future plant appearances is adequately feasible by multi-conditional CWGAN. The presented framework complements process-based models and overcomes their limitations, such as the reliance on assumptions and the low exact field-localization specificity, by realistic visualizations of the spatial crop development that directly lead to a high explainability of the model predictions.

Underwater image restoration based on progressive guidance

Underwater images often suffer from local distortions during the imaging and transmission process, which can negatively impact their quality. Fortunately, it is possible to improve image quality by removing local distortion without making any hardware or software adjustments to the transmission system. However, existing algorithms designed for global distortions are not suitable for addressing local distortions, while end-to-end restoration and inpainting algorithms do not perform satisfactorily on underwater images. To address this issue, this paper proposes a Joint distortion localization and restoration model based on Progressive Guidance (JPG) specifically tailored for underwater imaging and transmission. Our strategy employs a two-stage framework where the first stage focuses exclusively on accurately localizing distortions to obtain precise position. Subsequently, in the second stage, we utilize this position information for effective distortion restoration. To further enhance restoration performance, our approach progressively guides the restoration process by incorporating global, distortion-free as well as distortion-specific information into different components of the second-stage network. The work surpasses current state-of-the-art methods in restoring both mixed and individual distortions.

Local search resource allocation algorithm for space-based backbone network in Deep Reinforcement Learning method

With the evolution of Space-based backbone networks, the demand for enhanced efficiency and stability in network resource allocation has become increasingly critical, presenting a substantial challenge to conventional allocation methods. In response, we introduce an innovative resource allocation algorithm for space-based backbone networks. This algorithm represents a synergistic fusion of Deep Reinforcement Learning (DRL) and Local Search (LS) methodologies. It is specifically designed to reduce the extensive training duration associated with traditional policy networks, a crucial aspect in assuring optimal service quality. Our algorithm is structured within a two-stage framework that seamlessly integrates DRL and LS. A distinctive feature of our approach is the incorporation of link reliability into the algorithmic design. This element is meticulously tailored to address the dynamic and heterogeneous nature of space-based networks, ensuring effective resource management. The effectiveness of our approach is substantiated through extensive simulation results. These results demonstrate that the integration of DRL with LS not only enhances training efficiency but also exhibits significant improvements in resource allocation outcomes. Our work represents a noteworthy contribution to the development of practical optimization strategies in space-based networks, merging DRL with traditional methodologies for improved performance.

Unveiling the Terrain of Fraud: Exploring the Association Between Physical Topography and Consumer Fraud Victimization

ABSTRACT The widespread accessibility of the internet has facilitated social connections across geographical boundaries, but it has also led to an increase in consumer fraud victimization. This study aims to elucidate the geographical variations in fraud victimization by examining the influence of physical topography. Employing a two-stage conceptual framework of consumer fraud victimization, we used a nationwide survey in mainland China (n = 36,066) to explore the impact of mountainousness indicators on fraud exposure and fraud victimization. Our findings indicate a negative correlation between mean elevation and standard deviation in elevation with fraud exposure, while displaying a positive correlation with fraud victimization. Utilizing machine learning analyses, we discovered that mountainousness indicators outperformed factors such as income, physical condition, sex, education, and employment in predicting fraud victimization. Notably, physical topography played a more prominent role in the recognition model of fraud victimization compared to fraud exposure. These findings shed light on the fact that individuals residing in mountainous areas are less prone to fraud exposure but are at a heightened risk of becoming victims once exposed. This study offers valuable insights for the development of effective anti-fraud strategies.

Two-stage fusion framework driven by domain knowledge for penetration prediction of laser welding

To solve the problem of industrial application difficulties caused by insufficient depth exploration and poor interpretability of the pure data drive neural network. A two-stage fusion framework driven by domain knowledge was proposed and focused on industrial applications: laser welding penetration monitoring. First, a multi-sensor acquisition system based on plasma and molten pool signals is developed, extracting multiple features from plasma and molten pool images. Two Back Propagation (BP) neural network information fusion models are established in the first fusion stage, leveraging the characteristics of the plasma plume and molten pool front. In the second fusion stage, the Dempster-Shafer (DS) evidence theory is employed to fuse the information from the two BP neural network models. The proposed method, achieving a high accuracy of 98.7 %, is validated by comparing its average accuracy and anti-interference capability with classical Convolutional Neural Network (CNN) and BP neural network models. The physical relationship between input and output is elucidated through an established physical model, enhancing the interpretability and anti-interference capability of the model, and thereby improving the stability and accuracy of the system.