Prediction Process Research Articles

Risk Assessment of Multi-Hazards in Hangzhou: A Socioeconomic and Risk Mapping Approach Using the CatBoost-SHAP Model

As the global push for sustainable urban development progresses, this study, set against the backdrop of Hangzhou City, one of China’s megacities, addressed the conflict between urban expansion and the occurrence of urban geological hazards. Focusing on the predominant geological hazards troubling Hangzhou—urban road collapse, land subsidence, and karst collapse—we introduced a Categorical Boosting-SHapley Additive exPlanations (CatBoost-SHAP) model. This model not only demonstrates strong performance in predicting the selected typical urban hazards, with area under the curve (AUC) values reaching 0.92, 0.92, and 0.94, respectively, but also, through the incorporation of the explainable model SHAP, visually presents the prediction process, the interrelations between evaluation factors, and the weight of each factor. Additionally, the study undertook a multi-hazard evaluation, producing a susceptibility zoning map for multiple hazards, while performing tailored analysis by integrating economic and population density factors of Hangzhou. This research enables urban decision makers to transcend the “black box” limitations of machine learning, facilitating informed decision making through strategic resource allocation and scheduling based on economic and demographic factors of the study area. This approach holds the potential to offer valuable insights for the sustainable development of cities worldwide.

Detection of network false information based on artificial intelligence models

There is often some false information in social platforms to mislead public opinion. Due to the rapid development of the Internet, the spread of false information on the Internet has become easier, which has brought many losses to people's economy and life. In this paper, the relevant research on false information based on bidirectional convolutional networks is analyzed, and the method is divided into four stages: data preprocessing, model architecture, training process and prediction process. Then, the relevant research on rumor detection by propagation tree kernel model are analyzed. Finally, this paper delineates a comprehensive framework that amalgamates enhanced Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and a hybridized Black Widow Optimization (BWO) with Moth Optimization Algorithm (MOA) (referred to as HM-BWO) for the accurate detection of network-based false information. This paper analyzes and discusses the challenges of poor universality and insufficient detection speed encountered by the current rumor detection research and puts forward the idea of introducing migration model to solve the problem of poor universality and Spark to solve the problem of insufficient detection speed. This article provides a good overview of the field of online disinformation.

OPIN-ITP: Optimized Physics Informed Network with Trimmed Score Regression Based Insider Threats Prediction in Cloud Computing

INTRODUCTION: Insider threats are a major issue for cyber security. In contrast to external attackers, insiders have more privileges and authorized access to data and resources, which can cause an organization great harm. To completely understand an insider's activities throughout the organization, a more sophisticated method is needed. OBJECTIVES: Based on an organization's login activity, this study proposes a novel conceptual method for insider threat detection. Behavioural activities such as HTTP, Email and Login details are collected to create a dataset which is further processed for pre-processing using data transformation and Trimmed Score Regression (TSR). METHODS: These pre-data are given to the feature extraction process using Deep Feature Synthesis (DFS) extraction. The extracted data are fed to Physics Informed Neural Networks (PINN) for insider threat detection. RESULTS: The prediction process of PINN was improved through optimally choosing parameters such as learning rate and weight using Hunter-prey Optimization (HPO). The proposed model offers 68% detection rate, 98.4% accuracy, 5% FDR, 95% F1_score and 0.7005 sec execution time. CONCLUSION: Observed outcomes are compared to other traditional approaches of validation. The contrast with traditional approaches shows that the proposed model provides better outcomes than in traditional models and is therefore a good fit for real-time threat prediction.

Data-driven axial bearing capacity analysis of steel tubes infilled with rubberized alkali-activated concrete

This study aims to employ machine learning algorithms to analyze the axial bearing capacity of rubberized alkali-activated concrete filled steel tubes. A dataset encompassing 327 synthesized instances and seven input features is adopted for training and testing six machine learning models, including Decision Tree, Random Forest, Extremely Randomized Trees, Adaptive Boosting, Gradient Boosting Decision Trees (GBDT), and eXtreme Gradient Boosting Trees (XGBoost). The SHapley Additive exPlanation algorithm is employed to elucidate the prediction process of machine learning models and to analyze the influence of each parameter on axial bearing capacity. Comparison of evaluating metrics shows that GBDT and XGBoost models achieve highest accuracy and generalization capabilities when their Coefficient of Determination values surpassing 0.98 and Mean Absolute Percent Error remaining below 3%. Moreover, the explanation analysis of machine learning models reveals that diameter/width of the cross section, rubber content, yielding strength and thickness of steel tubes are critical variables that affect the axial bearing capacity, while compressive strength of alkali-activated concrete, specimen height, and shape of cross section show negligible impact. Besides, GBDT model overemphasizes the effect of specimen height and might lead a conservative prediction for specimens with smaller heights. Finally, compressive strength of alkali-activated concrete and diameter/width, thickness, and yielding strength of steel tubes are positively correlated with axial bearing capacity, and the increase of rubber content in alkali-activated concrete leads to the decrease of capacity.

Ultra-low cycle fatigue life prediction of stainless steel based on transfer learning guided artificial neural network

Determining ultra-low cycle fatigue (ULCF) life of stainless steel typically involves laborious and time-consuming tests. While machine learning offers an efficient solution for fatigue life prediction, the inherent demand for sufficient training data remains unaddressed. To solve this challenge, a novel method to predict ULCF life using small datasets was proposed by combining artificial neural network (ANN) and transfer learning (TL). A TL-guided ANN (TLNN) framework was developed, wherein the knowledge gained from a pre-trained ANN model for structural steels was transferred to ULCF life prediction of stainless steel. The TLNN model exhibits enhanced predictive capabilities for stainless steel under small amounts of training datasets, with mean R2 (coefficient of determination) value of 0.88 and average MAPE (mean absolute percentage error) value of 25.52%. In comparison, the directly trained ANN model shows lower performance, possessing that the average R2 and MAPE are 0.69 and 47.58%, respectively. We pioneeringly explored and found that adding synthetic data enhancement had no positive effect on the predictive ability of the TLNN model. Furthermore, the dependence of predictive capacity by the TLNN model on the number of training data availability was studied. The prediction process of the TLNN model was interpreted by SHAP (SHapley Additive exPlanations) method. In general, the developed TLNN model can efficiently predict the ULCF life of stainless steel with high accuracy and reduce the cost of consecutive fatigue property evaluation.

Enhancing soil liquefaction risk assessment with metaheuristics and hybrid learning techniques

ABSTRACT Soil liquefaction has garnered significant research attention over several decades due to its profound impact on infrastructure systems in earthquake-prone regions. Given the limitations inherent in the assumptions and approximations of conventional methods, machine learning has emerged as a robust and efficient approach for predicting soil liquefaction potential. This study aims to develop a synergistic JS-CNN-XGB model, unifying the Jellyfish Search (JS) optimizer with a hybrid deep/machine learning model. This amalgamation leverages the feature extraction capabilities of the Convolutional Neural Network (CNN) alongside the classification prowess of the eXtreme Gradient Boosting (XGB) algorithm. The proposed model seamlessly integrates into a user-friendly prediction system, streamlining the liquefaction prediction process. Validation is achieved through case studies involving historical earthquake recordings with diverse classification ratios and varying input attribute quantities. Compared to existing studies, our proposed system showcases a notable 6.2% enhancement in overall liquefaction assessment accuracy, demonstrating improved detection capabilities in imbalanced and balanced datasets. In conclusion, this study underscores an automated system that presents a robust solution to the challenges posed by liquefaction potential assessment in geotechnical engineering.

A predictive model for classifying college students' academic performance based on visual-spatial skills.

As the application of visual-spatial skills in academic disciplines, vocational fields and daily life is becoming more and more prominent, it is of great theoretical and practical significance how to make use of big data and artificial intelligence technology to conduct research on the relationship between visual-spatial skills and students' grades. This paper explores and analyses from the perspective of artificial intelligence, combining students' visual-spatial skills and students' specific attribute characteristics to construct an expert system, which defines the prediction of academic performance as a classification problem corresponding to the five categories of excellent, good, moderate, passing, and weak, respectively, and based on which a deep neural network-based classification prediction model for students' performance is designed. The experimental results show that visual-spatial skills plays an important role in the professional learning of science and engineering students, while the classification model designed in this paper has high accuracy in the grade prediction process. This paper not only helps to fill the gaps in the current research field, but is also expected to provide scientific basis for educational practice and promote the development of the education field in a more intelligent and personalized direction.

Three-dimensional generative adversarial networks for turbulent flow estimation from wall measurements

Different types of neural networks have been used to solve the flow sensing problem in turbulent flows, namely to estimate velocity in wall-parallel planes from wall measurements. Generative adversarial networks (GANs) are among the most promising methodologies, due to their more accurate estimations and better perceptual quality. This work tackles this flow sensing problem in the vicinity of the wall, addressing for the first time the reconstruction of the entire three-dimensional (3-D) field with a single network, i.e. a 3-D GAN. With this methodology, a single training and prediction process overcomes the limitation presented by the former approaches based on the independent estimation of wall-parallel planes. The network is capable of estimating the 3-D flow field with a level of error at each wall-normal distance comparable to that reported from wall-parallel plane estimations and at a lower training cost in terms of computational resources. The direct full 3-D reconstruction also unveils a direct interpretation in terms of coherent structures. It is shown that the accuracy of the network depends directly on the wall footprint of each individual turbulent structure. It is observed that wall-attached structures are predicted more accurately than wall-detached ones, especially at larger distances from the wall. Among wall-attached structures, smaller sweeps are reconstructed better than small ejections, while large ejections are reconstructed better than large sweeps as a consequence of their more intense footprint.

HRMF-DRP: A next-generation solution for overcoming provisioning challenges in cloud environments

The cloud computing infrastructure is a distributed environment and the existing research works have some provisioning problems such as suboptimal resource utilization and high execution time. The Heterogeneity Resource Management Framework for Dynamic Resource Provisioning (HRMF-DRP) is proposed for focusing on task scheduling and workload management. This framework incorporates advanced algorithms for dataset preprocessing, task clustering, workload prediction, and dynamic resource provisioning. For data preprocessing, the real-world workload traces were captured from the Planet Lab dataset that are taken as input for the preprocessing stage. The data preprocessing is responsible for ensuring data quality and reliability by using different models like missing data handling, outlier detection and removal as well as standardization and normalization. In this paper, the tasks are grouped into clusters by utilizing Density-Based Spatial Clustering of Applications with Noise (DBSCAN) model and this model categorizes the data points into border points, core points and noise points based on their density. The temporal dependencies are captured for the workload prediction by using Long Short-Term Memory (LSTM) neural network model. A Gaussian Mixture Model (GMM) model is responsible for estimating the number of Virtual machines (VMs) present in the workload prediction process. The Self-Adaptive Genetic Algorithm (SAGA) is implemented for task mapping that adjusts the parameters to change workload patterns for contributing adaptability and robustness. The different experimental evaluations are conducted based on the task completion time, workload balance index, resource utilization efficiency and workload prediction accuracy. The proposed model achieved the workload prediction accuracy of 98.5%, cost of $89.6, execution time of 125ms, Task Completion Time (TCT) of 40ms, Workload Balance Index (WBI) of 0.96 and Resource Utilization Efficiency (RUE) of 0.93. The quantitative results collectively position HRMF-DRP as a practical and efficient solution, promising advancements in dynamic resource provisioning for cloud computing, particularly within the Infrastructure as a Service (IaaS) cloud model.

A new financial risk prediction model based on deep learning and quasi-oppositional coot algorithm

Incorporating ground-breaking technologies such as deep learning (DL) has revolutionized predictive modelling in the rapidly evolving landscape of the finance sector. DL approaches, capable of extracting complex patterns from vast data collections, become an efficient approach for predicting financial trends. By integrating the complex neural network architecture with comprehensive datasets, including investor sentiment, market indicators, and economic variables, finance experts have introduced prediction models well known for their ability to capture the nuanced dynamics of financial markets with remarkable performance. Incorporating DL approaches within the finance sector provided the basis for more informed decision-making, enabling institutions, investors, and analysts to capitalize on emerging opportunities with greater confidence and precision and navigate market volatility. This study develops a novel quasi-oppositional coot algorithm with a deep learning-based predictive method on the financial sector (QOCODL-PMFC) technique. The QOCODL-PMFC technique aims to perform a prediction process on the financial sector. The QOCODL-PMFC method applies min-max normalization to measure the input dataset into a meaningful format to achieve this. Next, the QOCODL-PMFC method designs the QOCO technique for selecting an optimal set of features. The QOCODL-PMFC technique applies the attention bidirectional gated recurrent unit (ABiGRU) model for the prediction process. The Harris Hawks Optimization (HHO) model is utilized to boost the performance of the ABiGRU network. The simulation evaluation of the QOCODL-PMFC technique is tested under a benchmark finance dataset. The experimental values of the QOCODL-PMFC technique exhibit a minimal MSE of 0.7452 over other models.

Rock strength prediction based on machine learning: A study from prediction model to mechanism explanation

Rational and effective assessment of rock strength, especially uniaxial compressive strength (UCS), is the crucial basis of rock engineering practices. As computer technology advances, the application of machine learning techniques in predicting rock strength has become increasingly prevalent. However, due to the black-box effect of machine learning algorithms, the prediction process is always indescribable. Therefore, an innovative approach for the interpretable predicting uniaxial compressive strength (UCS) is initially introduced by using Explainable AI to crack the black-box effect of machine learning algorithms and explain the prediction mechanism. The findings reveal that the Xgboost model, optimized with interpretable hyper-parameters, achieves notable success in estimating the uniaxial compressive strength, with an RMSE of 14.69 and an R2 that can reach 0.9117. Additionally, multi-angle comparisons have verified that the robustness of the AOA-Xgboost model surpasses that of the other four hybrid models. The model employs SHAP analysis to measure how much each input factor contributes to the outcomes of the predictions. This work has potential for the prediction of crucial parameters and analysis of internal mechanisms of rock mass in geotechnical engineering.

Machine learning-based 3D scan coverage prediction for smart-control applications

Automatic control of a workpiece being manufactured is a requirement to ensure in-line correction and thus move towards a more intelligent manufacturing system. There is therefore a need to develop control strategies which are capable of taking precise account of real working conditions and enabling first-time-right control. As part of such a smart-control strategy, this paper introduces a machine learning-based approach capable of accurately predicting a priori the 3D coverage of a part according to a scan configuration given as input, i.e. predicting before scanning it which areas of the part will be acquired for real. This corresponds to a paradigm shift, where coverage estimation no longer relies on theoretical visibility criteria, but on rules learned from a large amount of data acquired in real-life conditions. The proposed 3D Scan Coverage Prediction Network (3DSCP-Net) is based on a 3D feature encoding and decoding module, which is capable of taking into account the specifics of the scan configuration whose impact on the 3D coverage is to be predicted. To take account of real working conditions, features are extracted at various levels, including geometric ones, but also features characterising the way structured-light projection behaves. The method is thus able to incorporate inter-reflection and overexposure issues into the prediction process. The database used for the training was built using an ad-hoc platform specially designed to enable the automatic acquisition and labelling of numerous point clouds from a wide variety of scan configurations. Experiments on several parts show that the method can efficiently predict the scan coverage, and that it outperforms conventional approaches based on purely theoretical visibility criteria.

OSTIS: A novel Organization-Specific Threat Intelligence System

With the increasing complexity and frequency of cyber attacks, organizations recognize the need for a proactive and targeted approach to safeguard their digital assets and operations. Every industry faces a distinct array of threats shaped by factors such as its industrial objective, geographic footprint, workforce size, revenue, partnerships, and the extent of its digital assets. This results in a wide heterogeneity in threat landscapes, which necessitates tailored threat intelligence sources. While some security practitioners may gravitate towards extensive sources, relying solely on volume-based solutions often leads to “alert fatigue”. For this reason, organization-specific threat intelligence has acquired a growing importance in cybersecurity defense.This work presents a complete and novel framework called OSTIS (Organization-Specific Threat Intelligence System) for generating and managing organization-specific Cyber Threat Intelligence (CTI) data. Our approach identifies reliable security blogs from which we gather CTI data through a custom and focused Web Crawler. Relevant content from such sources is, then, identified and extracted using automated deep-learning models. Moreover, our AI-driven solution maps CTI data to specific domain scenarios, such as education, finance, government, healthcare, industrial control systems, and IoT. To validate and gain insights from the trained models, we also include an explainable AI (XAI, for short) task carried out by leveraging the SHapley Additive exPlanations (SHAP) tool. This allows us to interpret the prediction process and discern influential content from data. The last step of our framework consists of the generation of an Organization Specific Threat Intelligence Knowledge Graph (OSTIKG), empowering organizations to identify and visualize attack patterns and incidents, promptly. To create this graph, we develop and adapt several techniques to extract diverse entities, including malware groups, campaigns, attack types, malware types, software tools, and so forth, and to identify relationships among them. Finally, through an extensive experimental campaign, we certify the validity and performance of all the components of our framework, which shows a 0.84 F1-score in the identification of relevant content, a 0.93 F1-score for the domain classification, and a 0.95 and 0.89 F1-score in the identification of entities and relations to build our OSTIKG graph.

To Study and Analyse the Customer Churn Prediction using Machine Learning Algorithm

The customer churn prediction (CCP) is one of the challenging problems in the E-Commerce industry. With the advancement in the field of machine learning and artificial intelligence, the possibilities to predict customer churn has increased significantly. Our proposed methodology, consists of six phases. In the first two phases, data pre-processing and feature analysis is performed. In the third phase, feature selection is taken into consideration. Next, the data has been split into two parts train and test set in the ratio of 80% and 20% respectively. In the prediction process, most popular predictive models have been applied, namely, logistic regression, random forest classifier etc. on train set are applied to see the effect on accuracy of models. In addition, K-fold cross validation has been used over train set for hyper parameter tuning and to prevent overfitting of models. Finally, the obtained results on test set have been evaluated using confusion matrix and AUC curve.

Hyperosmolar therapy response in traumatic brain injury: Explainable artificial intelligence based long-term time series forecasting approach

In the neurological intensive care setting implementing tier-based therapy, osmotic therapy serves as an intermediate approach for reducing intracranial pressure (ICP). However, discussions regarding quantitative and specific criteria for the maintenance and transition of treatment remain limited. This study proposes an explainable neural network forecasting of physiological responses after osmotic therapy administration for the purpose of treatment guidance. Recordings of arterial blood pressure and intracranial pressure were obtained from 107 patients with traumatic brain injury. To maintain adequate intracranial pressure, arterial blood pressure, and cerebral perfusion pressure, 20 % mannitol (100 ml; no. infusion, 3,571) and 11.3 % hypertonic saline (40 ml; no. infusion, 574) were administered. The LTSF-NLinear model was used to predict the target signal response for the subsequent 1 h. Six prediction models were derived by combining two agents (mannitol or hypertonic saline) and three target signals (mean arterial blood pressure, mean intracranial pressure and cerebral perfusion pressure). To validate the model’s robustness while considering patient heterogeneity, group (5)-fold cross-validation was implemented. To assess the predictive ability of the proposed model for adverse clinical events, binary classification of hypotension, intracranial hypertension, and cerebral hypoperfusion was performed. The overall function of the model was elucidated using linear weight visualization. Additionally, the DeepSHAP application allows for a detailed investigation of the prediction process for a single case. The relationship between response to osmotic therapy and patient outcomes was statistically significant. The proposed models achieved reasonable performance, with R2 score > 0.8 for physiological signal predictions and accuracy > 0.9 for event predictions. Through global explanation, the recent values of mean arterial blood pressure, mean intracranial pressure, and cerebral perfusion pressure of input signals significantly influenced physiological responses after osmotic treatment. In the context of local explanation (a single case), it was possible to determine how physiological state influenced the response. The proposed model is expected to provide objective and quantitative information on osmotherapy in neurological intensive care environments, enabling guidance for alternative and aggressive treatments. Thus, it can potentially aid in enhancing the prognosis of patients with traumatic brain injury and improving the clinical workflow. Moreover, the global and local explanations of the proposed model can provide valuable insights for future researchers.

Physics constrained pedestrian trajectory prediction with probability quantification

Accurately understanding and predicting the future trajectories of pedestrians around autonomous vehicles is a major challenge. Due to the uncertainty and diversity of pedestrian trajectories, current research focuses on the multimodality of trajectory prediction. Currently, multimodal pedestrian prediction methods mainly use deep generative networks to generate mutually independent trajectories through independent sampling strategy, thus focusing on modes with a large number of samples. Moreover, deep generative networks are black-box models that are entirely data-driven and cannot effectively explain trajectory generation. In addition, current research lacks an effective measure for each trajectory. Therefore, in our paper, we propose a physics constrained pedestrian trajectory prediction with probability quantification. We divide multimodal trajectory prediction into two phases: trajectory generation phase and trajectory selection phase. During the trajectory generation phase, rather than directly predicting trajectories, we incorporate a differential constraint module to ensure that the generated trajectories adhere to pedestrian motion laws. This approach enhances the interpretability of the model prediction process. Subsequently, we generate a set of candidate trajectory proposals with specific correlations through the correlated sampling module. During the trajectory selection phase, a probability selection module is proposed to establish explicit probability distributions for the candidate trajectory proposals and to select the proposal with the highest probability as the final output. Extensive experiments on a public real-world pedestrian trajectory dataset show that our proposed model exhibits significant advantages over existing models in multimodal trajectory prediction, which not only effectively reduces the error but also mitigates mode collapse. Moreover, we provide probabilities for each trajectory to better capture the model uncertainty and provide more comprehensive information for downstream decision-making systems.

Extract Implicit Semantic Friends and Their Influences from Bipartite Network for Social Recommendation

Social recommendation often incorporates trusted social links with user-item interactions to enhance rating prediction. Although methods that aggregate explicit social links have shown promising prospects, they are often constrained by the absence of explicit social data and the assumption of homogeneity, thus overlooking variations in social influence and consistency. These limitations hinder semantic expression and recommendation performance. Therefore, we propose a novel framework for social recommendation. First, we design a bipartite network embedding scheme, which learns vertex representations in the embedding space by modeling 1st-order explicit relations and higher-order implicit relations between vertices. Then, the similarity of the embedding vectors is used to extract top-k semantically consistent friends for each user. Next, we design an algorithm to assign a specific influence value to each user. Finally, we combine the top-k friends of the user and their influence values into an ensemble and add it as a regularization term to the rating prediction process of the user to correct the bias. Experiments on three real benchmark datasets show significant improvements in EISF over state-of-the-art methods.

Predicting surface roughness in machining aluminum alloys taking into account material properties

ABSTRACT This study investigates the use of machine learning models to predict surface roughness (Ra) in milling multi-grade aluminum alloys without prior knowledge of optimal cutting parameters. A diverse milling dataset encompassing material properties and cutting parameters from various aluminum alloy grades was compiled from research articles. Four machine learning algorithms, Extreme Gradient Boosting (XGB), Random Forest (RFR), Catalogical Gradient Boosting (CAT), and Gradient Boosting Regression (GBR), were employed to develop the predictive model. The dataset underwent cleaning, imputation, and outlier removal to ensure data quality. Feature engineering incorporated material properties and cutting parameters for model training. Performance metrics such as RMSE, MAPE, and R2 were used to assess the models’ accuracy. The SHapley Additive exPlanations (SHAP) technique was employed to interpret the models and identify influential features. GBR achieved the highest prediction accuracy with an RMSE of 0.2507 µm, MAPE of 23.36%, and R2 of 0.8709. Thermal conductivity, feed rate, and cutting speed were consistently identified as the most influential factors, although their rankings differed slightly. This study successfully developed a GBR model for effective Ra prediction in aluminum alloy milling, supporting advancements in smart manufacturing by enabling accurate surface quality prediction and data-driven process optimization through machine learning.

Fire resistance time prediction and optimization of cold-formed steel walls based on machine learning

Many full-scale experiments and numerical studies have been conducted to determine the fire performance of cold-formed steel (CFS) walls, but these studies are expensive and time consuming. This study proposes a machine learning (ML) based framework aiming at accurately predicting the fire resistance time (FRT) and optimizing the design of CFS walls under ISO 834 fire condition. To overcome the limitation of 32 experimental data points in the literature, a validated numerical method was used to generate 592 data points to expand the dataset of CFS walls, considering different wall configurations, various sheathing board types and thicknesses. The XGBoost (eXtreme Gradient Boosting) model was trained with numerical data and tested with experimental data. The hyperparameter tuning of the XGBoost model was implemented with Bayesian optimization, and it was found that the XGBoost model accurately predicted the FRT of CFS walls, with R2 and MAPE values being 0.933 and 8.46 %, respectively. The prediction process of the XGBoost model was interpreted by the SHAP (SHapley Additive exPlanations) method to determine the relative importance of input variables. The NSGA-II algorithm was adopted to optimize the FRT-cost dual-objective of CFS walls and the Pareto front including optimal solutions was emerging. The cost for each solution of Pareto front sets was lower than that in the real dataset at same FRT level. This phenomenon implied that the ML-based optimization framework successfully identified the cost-efficient design process. The proposed ML-based optimization framework offers a promising alternative for engineers to design CFS walls effectively with both mechanical and economic objectives.

Bedding-parallel fracture density prediction using graph convolutional network in continental shale oil reservoirs: A case study in Mahu Sag, Junggar basin, China

The natural fractures, e.g., the bedding-parallel fractures (BPF), have a significant impact on the storage space and horizontal permeability of continental shale oil reservoirs. However, the conventional logging response of BPF is complex. To solve the problem of BPF density prediction, we propose the Edge Generation Weighted Graph Convolutional Network (EG-WGCN) method, which is an improved Graph Convolutional Network (GCN). The new method integrates the relationship information between the fractures and lithology into the edge generation method, and adds the generated edge weight information into the message passing process, thereby effectively improving the accuracy of fracture density prediction. The prediction process is divided into three steps: first, using conventional logging sampling points as vertices, establish vertex sets for each lithology and single lithologic layer. Second, based on the depth sequence of a single lithologic layer and the Euclidean distance of the same lithologic vertex sets, the connection between vertices is established and overlapping edges generated by the above methods are given higher weights. Third, the vertices in the constructed graph are classified through graph convolutional layers, which are integrated into the edge weight information. This method is applied to the fracture density prediction of continental shale oil reservoirs in the Mahu Sag of the Junggar Basin, western China. The results show that the accuracy of our proposed EG-WGCN method in testing data is 95.13%. Compared to the GCN (boundless) without using the edge generation method and EG-GCN with the edge generation method but without the weighted aggregation mechanism, model accuracy is improved by 14.48% and 1.22%, respectively. In summary, this method proves that if a proper ML model is used, conventional logging data can become valuable for predicting BPF density with high accuracy even when cores are missing.