Integration Of Expert Knowledge Research Articles

Human-CoMo: A Combination of Cognitive Knowledge Engineering and Data Modelling for Human-Cyber-Physical Systems in Intelligent Manufacturing

Many cyber-physical systems face the challenge of appropriately integrating domain-specific human expert knowledge into the cyber part to create a shared sphere of knowledge and intelligent interactions between humans and the semi-autonomous technical system. Cognitive engineering contributes methods and insights into higher-order cognition that help to embed human knowledge in an appropriate way. The original research introduces a novel transdisciplinary framework called Human-CoMo, which demonstrates a systematic modelling process, different human perspectives, and the integration of expert knowledge at multiple hierarchical levels. Fundamental principles inspired by human cognition, such as conceptual chunking and knowledge precision, are characterised. Furthermore, it is shown how knowledge hierarchies can be methodically reflected in appropriate data analysis and modelling levels for small and big data applications including artificial intelligence approaches. Combined knowledge- and data-based modelling approaches offer more flexibility to integrate the strengths of humans and technology in a complementary way. The cognitive foundations and their computational reflections are outlined for the technical example process electroplating from the field of materials and surface engineering. The possibilities and limitations of integrating human knowledge through formalisation and implications for future forms of human-machine interaction are discussed.

Conditioned Protein Structure Prediction

Deep-learning-based protein structure prediction has facilitated major breakthroughs in biological sciences. However, current methods struggle with alternative conformation prediction and offer limited integration of expert knowledge on protein dynamics. We introduce AFEXplorer, a generic approach that tailors AlphaFold predictions to user-defined constraints in coarse coordinate spaces by optimizing embedding features. Its effectiveness in generating functional protein conformations in accordance with predefined conditions is demonstrated through comprehensive examples. AFEXplorer serves as a versatile platform for conditioned protein structure prediction, bridging the gap between automated models and domain-specific insights. Published by the American Physical Society 2024

Advanced Risk Assessment for Deep Excavation in Karst Regions Using Improved Dempster–Shafer and Dynamic Bayesian Networks

This study presents a novel risk-assessment methodology for deep foundation pit projects in karst regions, aimed at enhancing project safety and decision-making processes. This approach amalgamates fuzzy dynamic Bayesian networks with a refined Dempster–Shafer (DS) evidence theory to tackle the intricate uncertainties present in such contexts. A comprehensive risk index system, derived from historical accident cases, relevant standards, and the literature, encompasses environmental, design, construction, and management factors. Initial probabilities for each risk factor are determined through the integration of expert knowledge and fuzzy theory. The enhanced Dempster–Shafer theory is utilized to fuse diverse information sources, culminating in a robust and dynamic risk evaluation model. This model leverages real-time monitoring data to dynamically assess and adjust risk levels throughout the construction process. The validation of the proposed method is demonstrated through a detailed case study of the Guangzhou Tangxi Section 1 deep foundation pit project, which effectively identified critical risk factors and facilitated proactive construction strategy adjustments. To further evaluate the reliability of the methodology, comparisons were made with three alternative methods, and applications were conducted on three additional deep foundation pit projects. These comparative analyses confirm the superior reliability and applicability of the proposed methodology across varied scenarios.

Research on risk assessment of coal and gas outburst during continuous excavation cycle of coal mine with dynamic probabilistic inference

Coal and gas outbursts are a major dynamic hazard in underground coal mining, and predicting these events is challenging due to their complex mechanisms. This study introduces an advanced multistate dynamic probabilistic system designed to evaluate the risk of outbursts prior to each excavation cycle. Through the development of a Bayesian network (BN), we establish a dynamic model that captures the complex interactions among essential variables. The novel integration of time slices facilitates the continuous assessment across successive mining cycles. This study tackles the challenge of maintaining network parameter reliability in the static prediction conditions of outburst-prone coal mines. Specifically, in order to overcome the asynchronism of all variables testing process, the expert knowledge integration of expectation-maximization parameter learning and fuzzy set theory (FST) is built inside the system. Finally, we present a case study illustrating the application of this system across fifty-one consecutive cycles, with regular updates to variable states, showcasing the enhanced predictive capability of Dynamic Bayesian Networks (DBN) over traditional BNs for ongoing risk evaluation in mining operations. Additionally, by integrating backward diagnosis with sensitivity analysis, our approach simplifies the identification of key risk factors, offering valuable insights for improving outburst prevention measures and ensuring the safety of coal mining activities.

Digital Twins in Agriculture and Forestry: A Review.

Digital twins aim to optimize practices implemented in various sectors by bridging the gap between the physical and digital worlds. Focusing on open-field agriculture, livestock farming, and forestry and reviewing the current applications in these domains, this paper reveals the multifaceted roles of digital twins. Diverse key aspects are examined, including digital twin integration and maturity level, means of data acquisition, technological capabilities, and commonly used input and output features. Through the prism of four primary research questions, the state of the art of digital twins, the extent of their achieved integration, and an overview of the critical issues and potential advancements are provided in the landscape of the sectors under consideration. The paper concludes that in spite of the remarkable progress, there is a long way towards achieving full digital twin. Challenges still persist, while the key factor seems to be the integration of expert knowledge from different stakeholders. In light of the constraints identified in the review analysis, a new sector-specific definition for digital twins is also suggested to align with the distinctive characteristics of intricate biotic and abiotic systems. This research is anticipated to serve as a useful reference for stakeholders, enhancing awareness of the considerable benefits associated with digital twins and promoting a more systematic and comprehensive exploration of this transformative topic.

Enhanced Multitask Learning for Hash Code Generation of Palmprint Biometrics.

This paper presents a novel multitask learning framework for palmprint biometrics, which optimizes classification and hashing branches jointly. The classification branch within our framework facilitates the concurrent execution of three distinct tasks: identity recognition and classification of soft biometrics, encompassing gender and chirality. On the other hand, the hashing branch enables the generation of palmprint hash codes, optimizing for minimal storage as templates and efficient matching. The hashing branch derives the complementary information from these tasks by amalgamating knowledge acquired from the classification branch. This approach leads to superior overall performance compared to individual tasks in isolation. To enhance the effectiveness of multitask learning, two additional modules, an attention mechanism module and a customized gate control module, are introduced. These modules are vital in allocating higher weights to crucial channels and facilitating task-specific expert knowledge integration. Furthermore, an automatic weight adjustment module is incorporated to optimize the learning process further. This module fine-tunes the weights assigned to different tasks, improving performance. Integrating the three modules above has shown promising accuracies across various classification tasks and has notably improved authentication accuracy. The extensive experimental results validate the efficacy of our proposed framework.

The expert's knowledge combined with AI outperforms AI alone in seizure onset zone localization using resting state fMRI.

We evaluated whether integration of expert guidance on seizure onset zone (SOZ) identification from resting state functional MRI (rs-fMRI) connectomics combined with deep learning (DL) techniques enhances the SOZ delineation in patients with refractory epilepsy (RE), compared to utilizing DL alone. Rs-fMRI was collected from 52 children with RE who had subsequently undergone ic-EEG and then, if indicated, surgery for seizure control (n = 25). The resting state functional connectomics data were previously independently classified by two expert epileptologists, as indicative of measurement noise, typical resting state network connectivity, or SOZ. An expert knowledge integrated deep network was trained on functional connectomics data to identify SOZ. Expert knowledge integrated with DL showed a SOZ localization accuracy of 84.8 ± 4.5% and F1 score, harmonic mean of positive predictive value and sensitivity, of 91.7 ± 2.6%. Conversely, a DL only model yielded an accuracy of <50% (F1 score 63%). Activations that initiate in gray matter, extend through white matter, and end in vascular regions are seen as the most discriminative expert-identified SOZ characteristics. Integration of expert knowledge of functional connectomics can not only enhance the performance of DL in localizing SOZ in RE but also lead toward potentially useful explanations of prevalent co-activation patterns in SOZ. RE with surgical outcomes and preoperative rs-fMRI studies can yield expert knowledge most salient for SOZ identification.

Towards Fusing Data and Expert Knowledge for Better-Informed Digital Twins: An Initial Framework

Data-driven Digital Twin approaches have gained popularity in research literature, particularly in relation to Industry 4.0 achievements. Data-driven Digital Twins process widely available real-time data in an automated manner to extract models from various systems of interest. However, these model extraction approaches commonly lack systematic integration of expert knowledge from human experts, such as engineers. Combining valuable expert knowledge with data-driven model extraction approaches is a highly complex task that can significantly contribute to more accurate models within shorter time periods. In this paper, we provide a comprehensive overview of the research done on fusion of data and expert knowledge for Digital Twins to identify and describe the current gap. Resulting from our findings, we propose an initial framework for integrating expert knowledge and data for Digital Twin model extraction. Subsequently, we provide an overview of the main challenges and opportunities in fusing data and expert knowledge in the context of Digital Twins.

Large-scale group-hierarchical DEMATEL method for complex systems.

Existing Decision-Making Trial and Evaluation Laboratory (DEMATEL) methods are mostly suitable for simple systems with fewer factors, and lack effective integration of expert knowledge and experience from large-scale group populations, resulting in a potential compromise of the quality of the initial direct relation (IDR) matrix. To make DEMATEL better suited for the identification of critical factors in complex systems, this paper proposes a hierarchical DEMATEL method for large-scale group decision-making. Considering the limitations of expert knowledge and experience, a method based on expert consistency network for constructing the expert weight matrix is designed. The expert consistency network is constructed for different elements, and the weights of experts in different elements are determined using the clustering coefficient. Following the principles of the classic DEMATEL method, the steps for identifying key elements in complex systems using the large-scale group-hierarchical DEMATEL method are summarized. To objectively test the effectiveness and superiority of the decision algorithm, the robustness of the algorithm is analyzed in an interference environment. Finally, the superiority of the proposed method and algorithm is verified through a case study, which demonstrating that the proposed decision-making method is suitable for group decision-making in complex systems, with high algorithm stability and low algorithm deviation.

Machine learning enhanced characterization of surface defects using ultrasonic Rayleigh waves

Rayleigh waves are successfully employed for the characterization of surface defects through the utilization of both time and frequency domain analysis. However, a significant challenge arises when attempting to cover a broad range of the defect depth to Rayleigh wavelength ratio with a single method. This paper presents an enhanced methodology that integrates optimized machine learning (ML) algorithms to expand the coverage of defect depth assessment. An experimentally validated numerical model is firstly established to simulate the interaction of Rayleigh waves with surface defects and utilized to generate a significant quantity of labeled Rayleigh wave signals. The features of received Rayleigh wave patterns are effectively extracted and serve as indicators of defect parameters. The ML framework encompasses the extraction and selection of optimal features, the construction of model architecture, and the tuning of hyperparameters. Three dimensionality optimization techniques—principal component analysis, correlation coefficient analysis, and expert knowledge integration—are compared. Additionally, six machine learning models (decision tree, support vector machine, k-nearest neighbors, random forest, gradient-boosting machine, and artificial neural network) are optimized using particle swarm optimization and subjected to comparison. The results reveal that: (1) principal component analysis yields the highest performance, suggesting the potential for feature dimensionality reduction without relying on expert knowledge; (2) among the models examined, the random forest model and artificial neural network with hyperparameter tuning demonstrate superior performance in predicting defect depth; and (3) the optimized machine learning methods consistently outperform empirical approaches in accurately predicting defect depths across a wide range. Furthermore, the proposed method shows promise for extension to determine additional parameters associated with surface defects, including their width and angle of inclination.

FUCOM-optimization based predictive maintenance strategy using expert elicitation and Artificial Neural Network

Necessity of predictive maintenance appears when production assets such as machines need either a repair or a regular check. When it comes to monitoring digital twins of objects, maintenance predictions in time play a significant role. This study targets an artificial intelligence-based prediction of time-series data integrated with fully consistent expert elicitation. It also allows the integration of expert knowledge into the construction and explanation of the predictor. Moreover, it is considered a hybrid approach using data and expert knowledge for time series prediction. Reliability of obtained prediction is examined through comparisons with basic dynamic and static forecasts. Decision-making analysis based on expert’s opinions and optimization science coincides to prepare the data set for ANN prediction. Thus, the maintenance process will become predictable over time. Moreover, this prediction avoids sudden overhaul cost. Therefore, the entire production system is kept controlled. In this research, a FUll COnsistently Method (FUCOM) of optimization together with Multi Criteria Decision Making (MCDM) is used for obtaining optimal weights. Then, dynamic and static time-series forecasts are compared with ANN prediction. The obtained results prove ANN reliability according to expert’s opinions and FUCOM optimization model. FUCOM-ANN prediction achieves nearly 10 times smaller RMSE (0.08) than the RMSE (0.83) of general ANN prediction on the test data set. Moreover, the 10-fold cross-validation strategy shows better RMSE (0.13) for FUCOM-ANN validation sets than the RMSE (0.78) of general ANN validation sets.

A Review of Thermal Spectral Imaging Methods for Monitoring High-Temperature Molten Material Streams.

Real-time closed-loop control of metallurgical processes is still in its infancy, mostly based on simple models and limited sensor data and challenged by extreme temperature and harsh process conditions. Contact-free thermal imaging-based measurement approaches thus appear to be particularly suitable for process monitoring. With the potential to generate vast amounts of accurate data in real time and combined with artificial intelligence methods to enable real-time analysis and integration of expert knowledge, thermal spectral imaging is identified as a promising method offering more robust and accurate identification of key parameters, such as surface temperature, morphology, composition, and flow rate.

OOPS, the Ontology for Odor Perceptual Space: From Molecular Composition to Sensory Attributes of Odor Objects.

When creating a flavor to elicit a specific odor object characterized by odor sensory attributes (OSA), expert perfumers or flavorists use mental combinations of odor qualities (OQ) such as Fruity, Green, and Smoky. However, OSA and OQ are not directly related to the molecular composition in terms of odorants that constitute the chemical stimuli supporting odor object perception because of the complex non-linear integration of odor mixtures within the olfactory system. Indeed, single odorants are described with odor descriptors (OD), which can be found in various databases. Although classifications and aroma wheels studied the relationships between OD and OQ, the results were highly dependent on the studied products. Nevertheless, ontologies have proven to be very useful in sharing concepts across applications in a generic way and to allow experts' knowledge integration, implying non-linear cognitive processes. In this paper, we constructed the Ontology for Odor Perceptual Space (OOPS) to merge OD into a set of OQ best characterizing the odor, further translated into a set of OSA thanks to expert knowledge integration. Results showed that OOPS can help bridge molecular composition to odor perception and description, as demonstrated in the case of wines.

A traceable process to develop Bayesian networks from scarce data and expert judgment: A human reliability analysis application

The present paper develops a Bayesian Belief Network (BBN) for quantification of aggravating actions, as outcomes of inappropriate decisions, to be integrated in probabilistic safety assessment (PSA) models (i.e., the so-called errors of commission, EOCs). The BBN connects analyst ratings on influencing factors to the error forcing impact of a specific scenario, supporting the CESA-Q method (the Quantification module of the Commission Error Search and Assessment method). While contributing to the quantification of EOCs, this paper presents a novel process for the quantification of the BBN parameters (the Conditional Probability Distributions, CPDs), striving for traceable integration of expert knowledge and (scarce) data, in the form of retrospective analyses of operational events involving EOCs. The process combines the functional interpolation method for populating CPDs and Bayesian updates to adjust the BBN response to the available evidence. A first, prior BBN is developed, then sequentially updated to adjust to two data sets. This allows some intermediate validation and puts forwards the steps for future BBN updates as new EOC events (or new analyst assessments) become available.

Data-Based Stakeholder Identification in Technical Change Management

The efficient and effective handling of technical changes in product and production is seen as an important factor for the long-term success of manufacturing companies. Within the associated processes, the engineering and manufacturing change management, the identification and involvement of all relevant stakeholders, i.e., departments and employees, plays an essential role. Overlooking relevant stakeholders can lead to unforeseen impacts, such as production stops or further necessary changes, and can cause unforseen increased costs. In particular, in large companies, this task is complex and error-prone due to the high number of changes and departments involved, as well as the abundant variety of changes that can take place. Therefore, this contribution introduces an approach for stakeholder identification in technical change management, which allows the automated identification of relevant stakeholders at the beginning of the reactive phases of the change management process. The approach describes all necessary steps from data preparation to the evaluation of the obtained classification models. It is based on a text-classification approach and focuses in particular on the additional integration of expert knowledge to increase model quality. The approach has been successfully applied in cooperation with a German automotive company, and the obtained model quality has been compared to an expert-based classification.

Prevention of Falls from Heights in Construction Using an IoT System Based on Fuzzy Markup Language and JFML

The main cause of fatal accidents in the construction sector are falls from height (FFH) and the inappropriate use of a harness is commonly associated with these fatalities. Traditional methods, such as onsite inspections, safety communication, or safety training, are not enough to mitigate accidents caused by FFH associated with a poor management in the use of a harness. Although some technological solutions for the automated monitoring of workers could improve safety conditions, their use is not frequent due to the particularities of construction sites: complexity, dynamic environments, outdoor workplaces, etc. Then, the integration of expert knowledge with technology is a key issue. Fuzzy logic systems (FLS) and Internet of Things (IoT) present many potential benefits, such as real-time decisions being made based on FLS and data from sensors. In the current research, the development and test of an IoT system integrated with the Java Fuzzy Markup Language Library for FLS, to support experts’ decision making in FFH, is proposed. The proposal was checked in four construction scenarios based on working conditions with different levels of risk of FFH and obtained promising results.

Solving the genetic aetiology of hereditary gastrointestinal tumour syndromes– a collaborative multicentre endeavour within the project Solve-RD

BackgroundPatients and families with suspected, but genetically unexplained (unsolved) genetic tumour risk syndromes lack appropriate treatment and prevention, leading to preventable morbidity and mortality. To tackle this problem, patients from the European Reference Network on Genetic Tumour Risk Syndromes (ERN GENTURIS) are analysed in the European Commission's research project “Solving the unsolved rare diseases” (Solve-RD). The aim is to uncover known and novel cancer predisposing genes by reanalysing available whole-exome sequencing (WES) data of large cohorts in a combined manner, and applying a multidimensional omics approach. ApproachAround 500 genetically unsolved cases with suspected hereditary gastrointestinal tumour syndromes (polyposis, early-onset/familial colorectal cancer and gastric cancer) from multiple European centres are aimed to be included. Currently, clinical and germline WES data from 294 cases have been analysed. In addition, an extensive molecular profiling of gastrointestinal tumours from these patients is planned and deep learning techniques will be applied. The ambitious, multidisciplinary project is accompanied by a number of methodical, technical, and logistic challenges, which require the development and implementation of new analysis tools, the standardisation of bioinformatics pipelines, and strategies to exchange data and knowledge. Resultsand Outlook. The first re-analysis of 229 known and proposed cancer predisposition genes allowed solving 2–3% of previously unsolved GENTURIS cases. The integration of expert knowledge and new technologies will help to identify the genetic basis of additional unsolved cases within the ongoing project. The ERN GENTURIS approach might serve as a model for other genomic initiatives.

Integrating Expert Knowledge With Domain Adaptation for Unsupervised Fault Diagnosis

Data-driven fault diagnosis methods often require abundant labeled examples for each fault type. On the contrary, real-world data is often unlabeled and consists of mostly healthy observations and only few samples of faulty conditions. The lack of labels and fault samples imposes a significant challenge for existing data-driven fault diagnosis methods. In this paper, we aim to overcome this limitation by integrating expert knowledge with domain adaptation in a synthetic-to-real framework for unsupervised fault diagnosis. Motivated by the fact that domain experts often have a relatively good understanding on how different fault types affect healthy signals, in the first step of the proposed framework, a synthetic fault dataset is generated by augmenting real vibration samples of healthy bearings. This synthetic dataset integrates expert knowledge and encodes class information about the fault types. However, models trained solely based on the synthetic data often do not perform well because of the distinct distribution difference between the synthetically generated and real faults. To overcome this domain gap between the synthetic and real data, in the second step of the proposed framework, an imbalance-robust domain adaptation~(DA) approach is proposed to adapt the model from synthetic faults~(source) to the unlabeled real faults~(target) which suffer from severe class imbalance. The framework is evaluated on two unsupervised fault diagnosis cases for bearings, the CWRU laboratory dataset and a real-world wind-turbine dataset. Experimental results demonstrate that the generated faults are effective for encoding fault type information and the domain adaptation is robust against the different levels of class imbalance between faults.

Learning from Scarce Information: Using Synthetic Data to Classify Roman Fine Ware Pottery.

In this article, we consider a version of the challenging problem of learning from datasets whose size is too limited to allow generalisation beyond the training set. To address the challenge, we propose to use a transfer learning approach whereby the model is first trained on a synthetic dataset replicating features of the original objects. In this study, the objects were smartphone photographs of near-complete Roman terra sigillata pottery vessels from the collection of the Museum of London. Taking the replicated features from published profile drawings of pottery forms allowed the integration of expert knowledge into the process through our synthetic data generator. After this first initial training the model was fine-tuned with data from photographs of real vessels. We show, through exhaustive experiments across several popular deep learning architectures, different test priors, and considering the impact of the photograph viewpoint and excessive damage to the vessels, that the proposed hybrid approach enables the creation of classifiers with appropriate generalisation performance. This performance is significantly better than that of classifiers trained exclusively on the original data, which shows the promise of the approach to alleviate the fundamental issue of learning from small datasets.

A Multi-Source Data Feature Fusion and Expert Knowledge Integration Approach on Lithium-Ion Battery Anomaly Detection

Abstract To meet voltage and capability needs, batteries are grouped into packs as power sources. Abnormal ones in a pack will lead to partial heating and reduced available life, so removing anomalies out during manufacturing is of great significance. The conventional methods to detect abnormal batteries mainly rely on grading systems and manual operations. Current data-driven methods use statistical, machine learning and neural network approaches, building models, then applying them on the unlabeled. However, both cannot make full use of multiple source data and expert knowledge. Therefore, how to use these multi-source data and knowledge to improve the effect of battery anomaly detection process has become a research focus. We put forward a data-driven multi-source data feature fusion and expert knowledge integration (FFEKI) network architecture that follows encoder-decoder structure with multiple integration units and a corresponding joint loss function. First, we collect multi-source data and obtain fusion features. Then, we refine filters from expert knowledge and transform them into neural network layers as components of integration units. By this way, supervisory knowledge is integrated into our network. We evaluate our scheme by sets of experiments comparing with most widely used approaches on real manufacturing data. Results show that FFEKI obtains a maximum 100% anomaly detection rate (ADR). Meanwhile, when the number of detection T is greater than the actual number of anomalies in the testing set, our method can achieve full ADR faster. It is concluded that the proposed FFEKI achieves effective performance on power lithium-ion battery anomaly detection.