Model Output Research Articles

Teaching WebAR development with integrated machine learning: a methodology for immersive and intelligent educational experiences

Augmented reality (AR) and machine learning (ML) are rapidly growing technologies with immense potential for transforming education. Web-based augmented reality (WebAR) provides a promising approach to delivering immersive learning experiences on mobile devices. Integrating machine learning models into WebAR applications can enable advanced interactive effects by responding to user actions, thus enhancing the educational content. However, there is a lack of effective methodologies to teach students WebAR development with integrated machine learning. This paper proposes a methodology with three main steps: (1) Integrating standard TensorFlow.js models like handpose into WebAR scenes for gestures and interactions; (2) Developing custom image classification models with Teachable Machine and exporting to TensorFlow.js; (3) Modifying WebAR applications to load and use exported custom models, displaying model outputs as augmented reality content. The proposed methodology is designed to incrementally introduce machine learning integration, build an understanding of model training and usage, and spark ideas for using machine learning to augment educational content. The methodology provides a starting point for further research into pedagogical frameworks, assessments, and empirical studies on teaching WebAR development with embedded intelligence.

Stochastic representation and conditioning of process-based geological model by deep generative and recognition networks

Accurate and realistic geological modelling is the core of oil and gas development and production. In recent years, process-based methods are developed to produce highly realistic geological models by simulating the physical processes that reproduce the sedimentary events and develop the geometry. However, the complex dynamic processes are extremely expensive to simulate, making process-based models difficult to be conditioned to field data. In this work, we propose a comprehensive generative adversarial network framework as a machine-learning-assisted approach for mimicking the outputs of process-based geological models with fast generation. The main objective of our work is to obtain a continuous parametrization of the highly realistic process-based geological models which enables us to calibrate the models and condition the models to data. Numerical results are presented to illustrate the capability of our proposed methodology.

FORECASTING TECHNOLOGY ADOPTION BEHAVIOUR AND AGRICULTURAL PRODUCTION GROWTH

In this paper, we proposed a simple mathematical model for farmers in the Ranchi district to adopt and implement agricultural technologies. We used the Fisher-Pry model, which is a very effective model in the study of technological adoption. The model's output is a Sigmoid curve, or S-shaped curve, that develops exponentially at first, then approximately linearly, and lastly asymptotically. We have studied the models of technology spread using a basic case study and brought in a real-world situation. Mathematically, it is possible to predict that adoption will rise to the number m/2 when 50% of the farmers have adopted the technology, after which it will slow down. In a real-life situation, the point of inflection might occur before or after m/2.We also collected data on some of the food crops, and net irrigated area of Ranchi district from 2011 to 2020, and evaluated it graphically/mathematically. We examined the advantages of technology in agriculture using utility functions, by employing basic scenarios and selecting between different technologies for adoption

Interaction between clinicians and artificial intelligence to detect fetal atrioventricular septal defects on ultrasound: how can we optimize collaborative performance?

Artificial intelligence (AI) has shown promise in improving the performance of fetal ultrasound screening in detecting congenital heart disease (CHD). The effect of giving AI advice to human operators has not been studied in this context. Giving additional information about AI model workings, such as confidence scores for AI predictions, may be a way of further improving performance. Our aims were to investigate whether AI advice improved overall diagnostic accuracy (using a single CHD lesion as an exemplar), and to determine what, if any, additional information given to clinicians optimized the overall performance of the clinician-AI team. An AI model was trained to classify a single fetal CHD lesion (atrioventricular septal defect (AVSD)), using a retrospective cohort of 121 130 cardiac four-chamber images extracted from 173 ultrasound scan videos (98 with normal hearts, 75 with AVSD); a ResNet50 model architecture was used. Temperature scaling of model prediction probability was performed on a validation set, and gradient-weighted class activation maps (grad-CAMs) produced. Ten clinicians (two consultant fetal cardiologists, three trainees in pediatric cardiology and five fetal cardiac sonographers) were recruited from a center of fetal cardiology to participate. Each participant was shown 2000 fetal four-chamber images in a random order (1000 normal and 1000 AVSD). The dataset comprised 500 images, each shown in four conditions: (1) image alone without AI output; (2) image with binary AI classification; (3) image with AI model confidence; and (4) image with grad-CAM image overlays. The clinicians were asked to classify each image as normal or AVSD. A total of 20 000 image classifications were recorded from 10 clinicians. The AI model alone achieved an accuracy of 0.798 (95% CI, 0.760-0.832), a sensitivity of 0.868 (95% CI, 0.834-0.902) and a specificity of 0.728 (95% CI, 0.702-0.754), and the clinicians without AI achieved an accuracy of 0.844 (95% CI, 0.834-0.854), a sensitivity of 0.827 (95% CI, 0.795-0.858) and a specificity of 0.861 (95% CI, 0.828-0.895). Showing a binary (normal or AVSD) AI model output resulted in significant improvement in accuracy to 0.865 (P < 0.001). This effect was seen in both experienced and less-experienced participants. Giving incorrect AI advice resulted in a significant deterioration in overall accuracy, from 0.761 to 0.693 (P < 0.001), which was driven by an increase in both Type-I and Type-II errors by the clinicians. This effect was worsened by showing model confidence (accuracy, 0.649; P < 0.001) or grad-CAM (accuracy, 0.644; P < 0.001). AI has the potential to improve performance when used in collaboration with clinicians, even if the model performance does not reach expert level. Giving additional information about model workings such as model confidence and class activation map image overlays did not improve overall performance, and actually worsened performance for images for which the AI model was incorrect. © 2024 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Assessing and segmenting salt-affected soils using in-situ EC measurements, remote sensing, and a modified deep learning MU-NET convolutional neural network

A study revealed that the Siwa Oasis faces high soil salinity, which negatively impacts agricultural areas and crop productivity, despite its significant economic and agricultural importance. The current research proposes an approach to detect and segment salinity and vegetation areas at Siwa Oasis, Egypt, by combining remote sensing and building a deep learning neural network model-based U-NET algorithm to detect salinity change areas, anticipate further degradation, and predict soil quality indicators. To locate changes among the available images, standard image improvement, classification, and change detection methods have been used. We applied a deep learning modified U-Net (MU-NET) algorithm to segment and produce salinity maps. The MU-NET architecture is a two-level nested U-structure merged with a residual U-block (RUb), which consists of an encoder and a decoder. We applied RUb, which consists of several layers and skip connections. Different combinations of the salinity and vegetation indices were added to the original image to improve segmentation precision. The model was validated and trained using actual data samples collected over a 10-year period from the Landsat 8 satellite, which can monitor and analyse present land cover changes. The dataset consisted of 91 OLI and TIRS spectral images. Each one consists of eleven bands with a spatial resolution of 30 m for bands 1 to 7 and 9. A field survey was used as the main source of data for comparing the proposed model's outputs to assess the error rate. The study region is experiencing an increase in soil salinity in all directions, particularly with regard to the spatial distribution of saline soils, not just the quantitative increase in salt-affected soils. These findings supported the acceleration of soil salinization and vegetation death. The proposed model achieved the highest performance results among the other models and literature and was based on applying method 12 using 13 image layers, with the highest accuracies of 91.27% and 90.83% for salinity and vegetation, respectively.

Measuring the Gains and Losses of Virtual Water Flows in China’s Coastal Areas

The virtual water flow behind product trade provides researchers with ideas to alleviate water problems in China’s coastal areas, with accompanying resource and economic implications. This paper adopts a multi-region input–output model to calculate virtual water flows in coastal areas and analyses resource benefits and economic benefits, by combining the water stress index and shadow price, to measure gains and losses of virtual water flow. This study shows that (1) China’s coastal areas depends on external water supplies; (2) virtual water flows between coastal and inland areas generated 38.26 billion m3 of net scarce water savings and CNY 31,751 billion of net economic benefits, indicating that coastal areas benefited from virtual water flows from both resource and economic perspectives; (3) virtual water flows among coastal provinces and cities caused 11.10 billion m3 of net scarce water losses, but generated CNY 9784 billion of net economic gains, indicating that a further intensification of water stress in coastal areas, but such a virtual water flow pattern was beneficial from an economic perspective. This paper reveals the resource and economic impacts of virtual water flow at the national scale and among coastal provinces and cities and further proposes suggestions for promoting the beneficial flow of virtual water in coastal areas.

How does the establishment of an Indonesian Environmentally Extended Input Output (EEIO) model pave the way for Indonesia’s carbon future?

How does the establishment of an Indonesian Environmentally Extended Input Output (EEIO) model pave the way for Indonesia’s carbon future?

Establishing operator trust in machine learning for enhanced reliability and safety in nuclear Power Plants

The advancement of safety and reliability in Nuclear Power Plants (NPP) is essential for ensuring the protection of human life, the environment, and the sustainable use of clean energy. The key factor to achieve this goal is enhancing operator performance, as their decisions play a crucial role during reactor accidents. In this sense, the integration of artificial intelligence (AI) with safety systems in NPP could provide suggestions and recommendations to the operator, leading to a quick and accurate response. However, the complexity of models like Convolutional Neural Networks (CNN), often considered as black boxes, presents challenges for operators to comprehend the decision-making process. This difficulty leads to a lack of trust in the model's output. Therefore, ensuring algorithm transparency becomes crucial to maintain a positive relationship between humans and AI technology. In this study, the necessary data were collected through the automation technique implemented in the (PCTRAN) software. Subsequently, CNN was employed to classify various design-based accidents (DBA), accompanied by transfer learning to optimize accuracy. Finally, Shapley Additive Explanation (SHAP) was used to provide outcome explainability. The results show the effectiveness of automation in reducing time consumption and optimizing hardware resource utilization. Among the pre-trained models, MobileNet exhibits superior performance by requiring a smaller dataset and less time, achieving the highest accuracy, and demonstrating an optimal balance of (correct and incorrect) data prediction within the classes. Furthermore, the use of Explainable AI (XAI) provides result transparency and clarifies the reasons behind CNN incorrect predictions.

Building Trustworthy Generative Artificial Intelligence for Diabetes Care and Limb Preservation: A Medical Knowledge Extraction Case.

Large language models (LLMs) offer significant potential in medical information extraction but carry risks of generating incorrect information. This study aims to develop and validate a retriever-augmented generation (RAG) model that provides accurate medical knowledge about diabetes and diabetic foot care to laypersons with an eighth-grade literacy level. Improving health literacy through patient education is paramount to addressing the problem of limb loss in the diabetic population. In addition to affecting patient well-being through improved outcomes, improved physician well-being is an important outcome of a self-management model for patient health education. We used an RAG architecture and built a question-and-answer artificial intelligence (AI) model to extract knowledge in response to questions pertaining to diabetes and diabetic foot care. We utilized GPT-4 by OpenAI, with Pinecone as a vector database. The NIH National Standards for Diabetes Self-Management Education served as the basis for our knowledge base. The model's outputs were validated through expert review against established guidelines and literature. Fifty-eight keywords were used to select 295 articles and the model was tested against 175 questions across topics. The study demonstrated that with appropriate content volume and few-shot learning prompts, the RAG model achieved 98% accuracy, confirming its capability to offer user-friendly and comprehensible medical information. The RAG model represents a promising tool for delivering reliable medical knowledge to the public which can be used for self-education and self-management for diabetes, highlighting the importance of content validation and innovative prompt engineering in AI applications.

PrCRS: a prediction model of severe CRS in CAR-T therapy based on transfer learning

BackgroundCAR-T cell therapy represents a novel approach for the treatment of hematologic malignancies and solid tumors. However, its implementation is accompanied by the emergence of potentially life-threatening adverse events known as cytokine release syndrome (CRS). Given the escalating number of patients undergoing CAR-T therapy, there is an urgent need to develop predictive models for severe CRS occurrence to prevent it in advance. Currently, all existing models are based on decision trees whose accuracy is far from meeting our expectations, and there is a lack of deep learning models to predict the occurrence of severe CRS more accurately.ResultsWe propose PrCRS, a deep learning prediction model based on U-net and Transformer. Given the limited data available for CAR-T patients, we employ transfer learning using data from COVID-19 patients. The comprehensive evaluation demonstrates the superiority of the PrCRS model over other state-of-the-art methods for predicting CRS occurrence. We propose six models to forecast the probability of severe CRS for patients with one, two, and three days in advance. Additionally, we present a strategy to convert the model's output into actual probabilities of severe CRS and provide corresponding predictions.ConclusionsBased on our findings, PrCRS effectively predicts both the likelihood and timing of severe CRS in patients, thereby facilitating expedited and precise patient assessment, thus making a significant contribution to medical research. There is little research on applying deep learning algorithms to predict CRS, and our study fills this gap. This makes our research more novel and significant. Our code is publicly available at https://github.com/wzy38828201/PrCRS. The website of our prediction platform is: http://prediction.unicar-therapy.com/index-en.html.

Mapping analytical methods between input–output economics and network science

AbstractThe input–output (IO) model can be used to examine the flow of products and services within an economy, resembling a network with industries as nodes and transactions as links. Diverging significantly from commonly studied networks such as social, protein, and power grids, IO networks exhibit intricate interconnectivity, involving weighted nodes and both directional and weighted links. This uniqueness necessitates careful consideration when applying complex network analysis techniques to IO systems. We critically review current complex network metrics and attempt to link them with existing IO approaches. Based on our assessment, certain network metrics, such as degree centrality and eigenvector centrality, have been explicitly integrated into the IO theory. In contrast, there exist metrics whose definitions and interpretations expand when applied in the context of IO analysis, including closeness and betweenness centrality. Additionally, network metrics are usually used to study topological features, identify key sectors, and construct novel metrics to study related issues. Network metrics used in IO analysis can identify important driver and transmission sectors in resource flow and environmental emission network, facilitating the development of targeted and reliable strategies. Besides, network metrics are used to quantify topological features and structural changes of the IO network which help strengthen the supply chain and mitigate both direct and indirect impacts of disruptions. Our ultimate goal is to establish connections and offer a roadmap for developing network‐based tools in IO analysis.

Optimal allocation of water resources in the middle route of south-to-north water diversion project based on multi-regional input-output model

Long-distance water diversion, as an essential degree to alleviate the imbalance of regional water resource distribution, has become a crucial part of the global efforts of water conservancy and water ecology. However, the way to ensure green use of water resources while adequately diverting water is an urgent hassle to be solved. Based on relevant data from 11 cities in the water receiving areas along the Middle Route of the South-to-North Water Diversion Project and Markowitz's portfolio model, this study selects nine independent and representative indicators from the economic, social, and ecological aspects. Using the benefits of these nine indicators in 2015–2021 as a basis, the water diversion allocation portfolio with the greatest comprehensive benefits under normal operation of the Middle Route of the South-to-North Water Diversion Project was calculated and evaluated, and the benefits were measured and verified using the input–output model. The results demonstrate that, taking Beijing as an example, the comprehensive benefits will reach the maximum when the water supplied for economic, social, and ecological aspects is 15.16%, 56.35%, and 28.49%, respectively. At this time, economic benefits scored 9.50, which provide the greatest impetus to promote ecological and social development in water-affected areas along the route. While ecological and social benefits, which aim to protect the ecosystem and improve livelihoods, have a lower ability to promote development in water-affected areas.

The performance of large language models on quantitative and verbal ability tests: Initial evidence and implications for unproctored high‐stakes testing

AbstractUnproctored assessments are widely used in pre‐employment assessment. However, widely accessible large language models (LLMs) pose challenges for unproctored personnel assessments, given that applicants may use them to artificially inflate their scores beyond their true abilities. This may be particularly concerning in cognitive ability tests, which are widely used and traditionally considered to be less fakeable by humans than personality tests. Thus, this study compares the performance of LLMs on two common types of cognitive tests: quantitative ability (number series completion) and verbal ability (use a passage of text to determine whether a statement is true). The tests investigated are used in real‐world, high‐stakes selection. We also examine the performance of the LLMs across different test formats (i.e., open‐ended vs. multiple choice). Further, we contrast the performance of two LLMs (Generative Pretrained Transformers, GPT‐3.5 and GPT‐4) across multiple prompt approaches and “temperature” settings (i.e., a parameter that determines the amount of randomness in the model's output). We found that the LLMs performed well on the verbal ability test but extremely poorly on the quantitative ability test, even when accounting for the test format. GPT‐4 outperformed GPT‐3.5 across both types of tests. Notably, although prompt approaches and temperature settings did affect LLM test performance, those effects were mostly minor relative to differences across tests and language models. We provide recommendations for securing pre‐employment testing against LLM influences. Additionally, we call for rigorous research investigating the prevalence of LLM usage in pre‐employment testing as well as on how LLM usage affects selection test validity.

Development of a fed-batch-scale anaerobic co-digestion control system based on multivariable output error state space and model predictive control

This paper describes a feeding control system for fed-batch-scale anaerobic co-digestion. The proposed system is based on the multivariable output error state space and uses model predictive control to regulate the biogas flow output of anaerobic co-digestion. This control system combines predictive modeling, receding-horizon optimization, and state feedback. An accurate linear model of anaerobic co-digestion is obtained through the multivariable output error state space method using experimental data, and the accuracy of the model is evaluated using a goodness-of-fit index and the root mean square error. After augmenting the model, a predictive control system is established. The system states are observed through a Kalman filter, and the constrained receding-horizon optimization problem is solved using quadratic programming. Finally, the stability of the control system is demonstrated from both open- and closed-loop perspectives. The control system is then constructed and used in actual experiments, with the target flow output set at 100 NmL/h. The final flow output indicates that the control system achieves good control performance. This research enables the biogas output of anaerobic co-digestion to be controlled from a linear perspective through a concise control algorithm with strong practical application prospects.

Self-Supervised 3-D Semantic Representation Learning for Vision-and-Language Navigation.

In vision-and-language navigation (VLN) tasks, most current methods primarily utilize RGB images, overlooking the rich 3-D semantic data inherent to environments. To rectify this, we introduce a novel VLN framework that integrates 3-D semantic information into the navigation process. Our approach features a self-supervised training scheme that incorporates voxel-level 3-D semantic reconstruction to create a detailed 3-D semantic representation. A key component of this framework is a pretext task focused on region queries, which determines the presence of objects in specific 3-D areas. Following this, we devise an long short-term memory (LSTM)-based navigation model that is trained using our 3-D semantic representations. To maximize the utility of these 3-D semantic representations, we implement a cross-modal distillation strategy. This strategy encourages the RGB model's outputs to emulate those from the 3-D semantic feature network, enabling the concurrent training of both branches to merge RGB and 3-D semantic data effectively. Comprehensive evaluations on both the R2R and R4R datasets reveal that our method significantly enhances performance in VLN tasks.

A Hybrid Strategy Improved SPEA2 Algorithm for Multi-Objective Web Service Composition

A Hybrid Strategy Improved SPEA2 Algorithm for Multi-Objective Web Service Composition

Uniaxial and multiaxial cyclic deformation behavior prediction of Z2CN18.10 austenitic stainless steel based on Transformer deep learning method

In this study, the experimental study of cyclic deformation behavior of Z2CN18.10 austenitic stainless steel under uniaxial, cross, rhombus and circular paths is investigated. Based on Transformer deep learning method, three models are built to predict the cyclic deformation behavior of the material. One is the single input–single output model, which is used to predict the uniaxial cyclic deformation behavior. The second is the double input–single output model, which is used to predict the biaxial cyclic deformation behavior. The third is the time series multi-step model, which is used to predict the stress evolution under all paths. The Cross-Entropy loss function is adopted to train the models and the rationality of its application in the regression model is deduced from the theoretical point of view. Transformer deep learning method is based on attention mechanism, which can focus on the historical effect of Z2CN18.10 stainless steel cyclic deformation and accurately describe the uniaxial and multiaxial cyclic deformation behavior of the material.

Quasi-analog operation for low photon flux detection with SiPM's standard and fast outputs

In this paper, we characterize the quasi-analog detection performance of silicon photomultiplier (SiPM) photodetector for optical wireless communications (OWC) at low photon flux signal intensity. In addition to analyzing the specificities of conventional anode-cathode standard output, it also highlights the unique capacitively coupled fast output feature of the off-the-shelf SiPM devices introduced by SensL, now ON Semiconductor (onsemi). Furthermore, the potential advantages of fast output are analyzed, and first propose the combined output model of standard and fast outputs. This combined output can be used to suppress ambient noise utilizing the bipolar characteristics of the fast output amplitude, e.g., the amplitude of the fast output can be up to ±30 mV under a useful signal light intensity of 94.1 nW, which can suppress the ambient noise amplitude of 30 mV and effectively improves signal-to-noise ratio. In addition, an approximate linear quantitative relationship between the standard output voltage amplitude and the SiPM chip received incident optical power was obtained, e.g., at different intensities of dark current and ambient current (1 nA and 2 nA), standard output voltage amplitudes of 90 mV and 50 mV correspond to optical powers of 50.5 nW and 21.3 nW or 34.7 nW and 5.7 nW. Finally, the potential performance of the proposed combined output schemes is confirmed by experimental and the measured Q-factor of the eye diagrams was optimised from 13.03 to 15.08. The proposed onsemi SiPM combined output schemes of standard and fast outputs can be selected to detection analog low-light link.

Decomposition of China’s Carbon Emissions Responsibility from the Perspective of Technological Heterogeneity

A global agreement has been reached on the reduction in greenhouse gas emissions. Worldwide, countries have implemented measures to tackle carbon emission issues by establishing aggregate targets and decomposing responsibilities. This study aims to decompose carbon emissions by creating an input–output model that incorporates multivariate factors like energy consumption and water consumption, together with a ZSG-DEA (zero-sum data envelopment analysis) model considering technological heterogeneity (Tech-ZSG-DEA). Based on the total carbon emission data predicted using the STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology) model, the carbon emission efficiency of 30 provinces in China is evaluated according to multiple principles. This is achieved by considering variations in population, economy, energy consumption, and water consumption across different locations. The efficiency findings indicate a discrepancy between the initial allocation and the highest efficiency value of 1. The traditional ZSG-DEA model overlooks regional disparities and may worsen carbon emission pressures in less developed areas. In contrast, the Tech-ZSG-DEA model, which considers regional technological diversity, allows more efficient regions to help alleviate some of the carbon emission burden and considers economic and social benefits. There is a large difference in the emission responsibility of the provinces based on the different decomposition principles. Finally, relevant policy recommendations are provided, such as the formulation of differentiated and inclusively coordinated emission plans. In addition, there are also mechanisms for coordinating interests and joint prevention among different regions.

Design of a large language model for improving customer service in telecom operators

AbstractTelecommunications operators are tasked with enhancing service quality, reducing operational costs, and preserving customer privacy. This study presents an innovative application of large language models (LLMs) integrated with the LangChain technology framework, aimed at revolutionizing customer service in the telecom sector. The LangChain framework features a Knowledge Organizing Module and a Knowledge Search Module, both designed to refine customer support operations. The research develops an LLM‐based approach to improve the segmentation and organization of knowledge bases, tailored for the telecommunications industry. This approach ensures seamless integration with existing LLMs while preserving distinct knowledge domains, crucial for search accuracy. Additionally, the framework includes an advanced information security protocol with a robust filtering system that effectively excludes sensitive data from the model's outputs, enhancing data security. Empirical findings indicate that the ChatGLM2‐6B+LangChain model outperforms the baseline ChatGLM2, demonstrating heightened effectiveness in telecom‐specific tasks and outstripping even more sophisticated models like GPT‐4. The implementation of this LLM‐based framework within telecom customer service systems has significantly sharpened the precision of knowledge recommendations, as reflected by a dramatic increase in acceptance rates from 15% to 70%.