Model Performance In Terms Research Articles

RDGT-GAN: Robust Distribution Generalization of Transformers for COVID-19 Fake News Detection

Social media platforms have become a vital source of information during the outbreak of the pandemic (COVID-19). The phenomena of fake information or news spread through social media have become increasingly prevalent and a powerful tool for information proliferation. Detecting fake news is crucial for the betterment of society. Existing fake news detection models focus on increasing the performance which leads to overfitting and lag generalizability. Hence, these models require training for various datasets of the same domain with significant variations in the distribution. In our work, we have addressed this overfitting issue by designing a robust distribution generalization of transformers-based generative adversarial network (RDGT-GAN) architecture, which can generalize the model for COVID-19 fake news datasets with different distributions without retraining. Based on our experimental findings, it is evident that the proposed model outperforms the current state-of-the-art (SOTA) models in terms of performance.

On optimization of calibrations of a distributed hydrological model with spatially distributed information on snow

Abstract. In northern cold-temperate countries, a large portion of annual streamflow is produced by spring snowmelt, which often triggers floods. It is important to have spatial information about snow variables such as snow water equivalent (SWE), which can be incorporated into hydrological models, making them more efficient tools for improved decision-making. The present research implements a unique spatial pattern metric in a multi-objective framework for calibration of hydrological models and attempts to determine whether raw SNODAS (SNOw Data Assimilation System) data can be utilized for hydrological model calibration. The spatial efficiency (SPAEF) metric is explored for spatially calibrating SWE. Different calibration experiments are performed combining Nash–Sutcliffe efficiency (NSE) for streamflow and root-mean-square error (RMSE) and SPAEF for SWE, using the Dynamically Dimensioned Search (DDS) and Pareto Archived Dynamically Dimensioned Search multi-objective optimization (PADDS) algorithms. Results of the study demonstrate that multi-objective calibration outperforms sequential calibration in terms of model performance (SWE and discharge simulations). Traditional model calibration involving only streamflow produced slightly higher NSE values; however, the spatial distribution of SWE could not be adequately maintained. This study indicates that utilizing SPAEF for spatial calibration of snow parameters improved streamflow prediction compared to the conventional practice of using RMSE for calibration. SPAEF is further implied to be a more effective metric than RMSE for both sequential and multi-objective calibration. During validation, the calibration experiment incorporating multi-objective SPAEF exhibits enhanced performance in terms of NSE and Kling–Gupta efficiency (KGE) compared to calibration experiment solely based on NSE. This observation supports the notion that incorporating SPAEF computed on raw SNODAS data within the calibration framework results in a more robust hydrological model. The novelty of this study is the implementation of SPAEF with respect to spatially distributed SWE for calibrating a distributed hydrological model.

Deep learning solutions for smart city challenges in urban development

In the realm of urban planning, the integration of deep learning technologies has emerged as a transformative force, promising to revolutionize the way cities are designed, managed, and optimized. This research embarks on a multifaceted exploration that combines the power of deep learning with Bayesian regularization techniques to enhance the performance and reliability of neural networks tailored for urban planning applications. Deep learning, characterized by its ability to extract complex patterns from vast urban datasets, has the potential to offer unprecedented insights into urban dynamics, transportation networks, and environmental sustainability. However, the complexity of these models often leads to challenges such as overfitting and limited interpretability. To address these issues, Bayesian regularization methods are employed to imbue neural networks with a principled framework that enhances generalization while quantifying predictive uncertainty. This research unfolds with the practical implementation of Bayesian regularization within neural networks, focusing on applications ranging from traffic prediction, urban infrastructure, data privacy, safety and security. By integrating Bayesian regularization, the aim is to, not only improve model performance in terms of accuracy and reliability but also to provide planners and decision-makers with probabilistic insights into the outcomes of various urban interventions. In tandem with quantitative assessments, graphical analysis is wielded as a crucial tool to visualize the inner workings of deep learning models in the context of urban planning. Through graphical representations, network visualizations, and decision boundary analysis, we uncover how Bayesian regularization influences neural network architecture and enhances interpretability.

TCDformer: A transformer framework for non-stationary time series forecasting based on trend and change-point detection

Although time series prediction models based on Transformer architecture have achieved significant advances, concerns have arisen regarding their performance with non-stationary real-world data. Traditional methods often use stabilization techniques to boost predictability, but this often results in the loss of non-stationarity, notably underperforming when tackling major events in practical applications. To address this challenge, this research introduces an innovative method named TCDformer (Trend and Change-point Detection Transformer). TCDformer employs a unique strategy, initially encoding abrupt changes in non-stationary time series using the local linear scaling approximation (LLSA) module. The reconstructed contextual time series is then decomposed into trend and seasonal components. The final prediction results are derived from the additive combination of a multilayer perceptron (MLP) for predicting trend components and wavelet attention mechanisms for seasonal components. Comprehensive experimental results show that on standard time series prediction datasets, TCDformer significantly surpasses existing benchmark models in terms of performance, reducing MSE by 47.36% and MAE by 31.12%. This approach offers an effective framework for managing non-stationary time series, achieving a balance between performance and interpretability, making it especially suitable for addressing non-stationarity challenges in real-world scenarios.

Semi-supervised auto-segmentation method for pelvic organ-at-risk in magnetic resonance images based on deep-learning.

In radiotherapy, magnetic resonance (MR) imaging has higher contrast for soft tissues compared to computed tomography (CT) scanning and does not emit radiation. However, manual annotation of the deep learning-based automatic organ-at-risk (OAR) delineation algorithms is expensive, making the collection of large-high-quality annotated datasets a challenge. Therefore, we proposed the low-cost semi-supervised OAR segmentation method using small pelvic MR image annotations. We trained a deep learning-based segmentation model using 116 sets of MR images from 116 patients. The bladder, femoral heads, rectum, and small intestine were selected as OAR regions. To generate the training set, we utilized a semi-supervised method and ensemble learning techniques. Additionally, we employed a post-processing algorithm to correct the self-annotation data. Both 2D and 3D auto-segmentation networks were evaluated for their performance. Furthermore, we evaluated the performance of semi-supervised method for 50 labeled data and only 10 labeled data. The Dice similarity coefficient (DSC) of the bladder, femoral heads, rectum and small intestine between segmentation results and reference masks is 0.954, 0.984, 0.908, 0.852 only using self-annotation and post-processing methods of 2D segmentation model. The DSC of corresponding OARs is 0.871, 0.975, 0.975, 0.783, 0.724 using 3D segmentation network, 0.896, 0.984, 0.890, 0.828 using 2D segmentation network and common supervised method. The outcomes of our study demonstrate that it is possible to train a multi-OAR segmentation model using small annotation samples and additional unlabeled data. To effectively annotate the dataset, ensemble learning and post-processing methods were employed. Additionally, when dealing with anisotropy and limited sample sizes, the 2D model outperformed the 3D model in terms of performance.

DeepSF-4mC: A deep learning model for predicting DNA cytosine 4mC methylation sites leveraging sequence features

N4-methylcytosine (4mC) is a DNA modification involving the addition of a methyl group to the fourth nitrogen atom of the cytosine base. This modification may influence gene regulation, providing potential insights into gene control mechanisms. Traditional laboratory methods for detecting 4mC DNA methylation have limitations, but the rise of artificial intelligence has introduced efficient computational strategies for 4mC site prediction. Despite this progress, challenges persist in terms of model performance and interpretability. To tackle these challenges, we propose DeepSF-4mC, a deep learning model specifically designed for predicting DNA cytosine 4mC methylation sites by leveraging sequence features. Our approach incorporates multiple encoding techniques to enhance prediction accuracy, increase model stability, and reduce the computational resources needed. Leveraging transfer learning, we harness existing models to enhance performance through learned representations or fine-tuning. Ensemble learning techniques combine predictions from multiple models, boosting robustness and accuracy. This research contributes to DNA methylation analysis and lays the groundwork for understanding 4mC's multifaceted role in biological processes. The web server for DeepSF-4mC is accessible at: http://deepsf-4mc.top/and the original code can be found at: https://github.com/754131799/DeepSF-4mC.

Towards cross-silo federated learning for corporate organizations

Digital media companies rely on machine learning models to target their content toward their audience’s interests. Machine learning models usually rely on the amount and quality of training data. While today, data is abundant, it is typically stored in data silos and cannot be shared between companies or publishers due to data protection and user privacy. Federated Learning (FL) is a distributed machine learning approach that is rapidly gaining popularity and enables collaboratively training machine learning models on a large corpus of decentralized data. Prior research on FL mainly focuses on an FL setup containing millions of clients. For example, a client may be a single user’s mobile device with data. However, we note that, in many scenarios, corporate organizations such as news media companies that have available data from multiple sets of users could also benefit from FL. In this work, we aim to focus on building FL models where multiple corporate organizations like news media companies or banks participate in the training process of FL to collaboratively train federated models. We used federated learning to train models for a set of corporate stakeholders and applied FL for two tasks: a classification task and a ranking task. For the classification task, we designed a tree-based federated random forest algorithm and a neural network-based federated algorithm. For the ranking task, we designed a federated neural ranking model for news article recommendations. Our experimental results demonstrate that corporate companies by participating in FL can achieve benefits in improving the model performance in terms of accuracy for classification tasks and in terms of ranking for recommendation tasks. Furthermore, we designed and developed a simple framework for a small number of stakeholders to train federated models.

IFGAN—A Novel Image Fusion Model to Fuse 3D Point Cloud Sensory Data

To enhance the level of autonomy in driving, it is crucial to ensure optimal execution of critical maneuvers in all situations. However, numerous accidents involving autonomous vehicles (AVs) developed by major automobile manufacturers in recent years have been attributed to poor decision making caused by insufficient perception of environmental information. AVs employ diverse sensors in today’s technology-driven settings to gather this information. However, due to technical and natural factors, the data collected by these sensors may be incomplete or ambiguous, leading to misinterpretation by AVs and resulting in fatal accidents. Furthermore, environmental information obtained from multiple sources in the vehicular environment often exhibits multimodal characteristics. To address this limitation, effective preprocessing of raw sensory data becomes essential, involving two crucial tasks: data cleaning and data fusion. In this context, we propose a comprehensive data fusion engine that categorizes various sensory data formats and appropriately merges them to enhance accuracy. Specifically, we suggest a general framework to combine audio, visual, and textual data, building upon our previous research on an innovative hybrid image fusion model that fused multispectral image data. However, this previous model faced challenges when fusing 3D point cloud data and handling large volumes of sensory data. To overcome these challenges, our study introduces a novel image fusion model called Image Fusion Generative Adversarial Network (IFGAN), which incorporates a multi-scale attention mechanism into both the generator and discriminator of a Generative Adversarial Network (GAN). The primary objective of image fusion is to merge complementary data from various perspectives of the same scene to enhance the clarity and detail of the final image. The multi-scale attention mechanism serves two purposes: the first, capturing comprehensive spatial information to enable the generator to focus on foreground and background target information in the sensory data, and the second, constraining the discriminator to concentrate on attention regions rather than the entire input image. Furthermore, the proposed model integrates the color information retention concept from the previously proposed image fusion model. Furthermore, we propose simple and efficient models for extracting salient image features. We evaluate the proposed models using various standard metrics and compare them with existing popular models. The results demonstrate that our proposed image fusion model outperforms the other models in terms of performance.

Earth system modeling on modular supercomputing architecture: coupled atmosphere–ocean simulations with ICON 2.6.6-rc

Abstract. The confrontation of complex Earth system model (ESM) codes with novel supercomputing architectures poses challenges to efficient modeling and job submission strategies. The modular setup of these models naturally fits a modular supercomputing architecture (MSA), which tightly integrates heterogeneous hardware resources into a larger and more flexible high-performance computing (HPC) system. While parts of the ESM codes can easily take advantage of the increased parallelism and communication capabilities of modern GPUs, others lag behind due to the long development cycles or are better suited to run on classical CPUs due to their communication and memory usage patterns. To better cope with these imbalances between the development of the model components, we performed benchmark campaigns on the Jülich Wizard for European Leadership Science (JUWELS) modular HPC system. We enabled the weather and climate model Icosahedral Nonhydrostatic (ICON) to run in a coupled atmosphere–ocean setup, where the ocean and the model I/O is running on the CPU Cluster, while the atmosphere is simulated simultaneously on the GPUs of JUWELS Booster (ICON-MSA). Both atmosphere and ocean are running globally with a resolution of 5 km. In our test case, an optimal configuration in terms of model performance (core hours per simulation day) was found for the combination of 84 GPU nodes on the JUWELS Booster module to simulate the atmosphere and 80 CPU nodes on the JUWELS Cluster module, of which 63 nodes were used for the ocean simulation and the remaining 17 nodes were reserved for I/O. With this configuration the waiting times of the coupler were minimized. Compared to a simulation performed on CPUs only, the MSA approach reduces energy consumption by 45 % with comparable runtimes. ICON-MSA is able to scale up to a significant portion of the JUWELS system, making best use of the available computing resources. A maximum throughput of 170 simulation days per day (SDPD) was achieved when running ICON on 335 JUWELS Booster nodes and 268 Cluster nodes.

GANDALF: Graph-based transformer and Data Augmentation Active Learning Framework with interpretable features for multi-label chest Xray classification

Informative sample selection in an active learning (AL) setting helps a machine learning system attain optimum performance with minimum labeled samples, thus reducing annotation costs and boosting performance of computer-aided diagnosis systems in the presence of limited labeled data. Another effective technique to enlarge datasets in a small labeled data regime is data augmentation. An intuitive active learning approach thus consists of combining informative sample selection and data augmentation to leverage their respective advantages and improve the performance of AL systems. In this paper, we propose a novel approach called GANDALF (Graph-based TrANsformer and Data Augmentation Active Learning Framework) to combine sample selection and data augmentation in a multi-label setting. Conventional sample selection approaches in AL have mostly focused on the single-label setting where a sample has only one disease label. These approaches do not perform optimally when a sample can have multiple disease labels (e.g., in chest X-ray images). We improve upon state-of-the-art multi-label active learning techniques by representing disease labels as graph nodes and use graph attention transformers (GAT) to learn more effective inter-label relationships. We identify the most informative samples by aggregating GAT representations. Subsequently, we generate transformations of these informative samples by sampling from a learned latent space. From these generated samples, we identify informative samples via a novel multi-label informativeness score, which beyond the state of the art, ensures that (i) generated samples are not redundant with respect to the training data and (ii) make important contributions to the training stage. We apply our method to two public chest X-ray datasets, as well as breast, dermatology, retina and kidney tissue microscopy MedMNIST datasets, and report improved results over state-of-the-art multi-label AL techniques in terms of model performance, learning rates, and robustness.

Effects of AST-120 on mortality in patients with chronic kidney disease modeled by artificial intelligence or traditional statistical analysis

Chronic kidney disease (CKD) imposes a substantial burden, and patient prognosis remains grim. The impact of AST-120 (AST-120) on the survival of CKD patients lacks a consensus. This study aims to investigate the effects of AST-120 usage on the survival of CKD patients and explore the utility of artificial intelligence models for decision-making. We conducted a retrospective analysis of CKD patients receiving care in the pre-end-stage renal disease (ESRD) program at Taichung Veterans General Hospital from 2000 to 2019. We employed Cox regression models to evaluate the relationship between AST-120 use and patient survival, both before and after propensity score matching. Subsequently, we employed Deep Neural Network (DNN) and Extreme Gradient Boosting (XGBoost) models to assess their performance in predicting AST-120's impact on patient survival. Among the 2584 patients in our cohort, 2199 did not use AST-120, while 385 patients received AST-120. AST-120 users exhibited significantly lower mortality rates compared to non-AST-120 users (13.51% vs. 37.88%, p < 0.0001) and a reduced prevalence of ESRD (44.16% vs. 53.17%, p = 0.0005). Propensity score matching at 1:1 and 1:2 revealed no significant differences, except for dialysis and all-cause mortality, where AST-120 users exhibited significantly lower all-cause mortality (p < 0.0001), with a hazard ratio (HR) of 0.395 (95% CI = 0.295–0.522). This difference remained statistically significant even after propensity matching. In terms of model performance, the XGBoost model demonstrated the highest accuracy (0.72), specificity (0.90), and positive predictive value (0.48), while the logistic regression model showed the highest sensitivity (0.63) and negative predictive value (0.84). The area under the curve (AUC) values for logistic regression, DNN, and XGBoost were 0.73, 0.73, and 0.69, respectively, indicating similar predictive capabilities for mortality. In this cohort of CKD patients, the use of AST-120 is significantly associated with reduced mortality. However, the performance of artificial intelligence models in predicting the impact of AST-120 is not superior to statistical analysis using the current architecture and algorithm.

Impact of random oversampling and random undersampling on the performance of prediction models developed using observational health data

BackgroundThere is currently no consensus on the impact of class imbalance methods on the performance of clinical prediction models. We aimed to empirically investigate the impact of random oversampling and random undersampling, two commonly used class imbalance methods, on the internal and external validation performance of prediction models developed using observational health data.MethodsWe developed and externally validated prediction models for various outcomes of interest within a target population of people with pharmaceutically treated depression across four large observational health databases. We used three different classifiers (lasso logistic regression, random forest, XGBoost) and varied the target imbalance ratio. We evaluated the impact on model performance in terms of discrimination and calibration. Discrimination was assessed using the area under the receiver operating characteristic curve (AUROC) and calibration was assessed using calibration plots.ResultsWe developed and externally validated a total of 1,566 prediction models. On internal and external validation, random oversampling and random undersampling generally did not result in higher AUROCs. Moreover, we found overestimated risks, although this miscalibration could largely be corrected by recalibrating the models towards the imbalance ratios in the original dataset.ConclusionsOverall, we found that random oversampling or random undersampling generally does not improve the internal and external validation performance of prediction models developed in large observational health databases. Based on our findings, we do not recommend applying random oversampling or random undersampling when developing prediction models in large observational health databases.

SiamTHN: Siamese Target Highlight Network for Visual Tracking

Siamese network based trackers develop rapidly in the field of visual object tracking in recent years. The majority of Siamese network based trackers now in use treat each channel in the feature maps generated by the backbone network equally, making the similarity response map sensitive to background influence and hence challenging to focus on the target region. Additionally, there are no structural links between the classification and regression branches in these trackers, and the two branches are optimized separately during training. Therefore, there is a misalignment between the classification and regression branches, which results in less accurate tracking results. In this paper, a Target Highlight Module is proposed to help the generated similarity response maps to be more focused on the target region. To reduce the misalignment and produce more precise tracking results, we propose a corrective loss to train the model. The two branches of the model are jointly tuned with the use of corrective loss to produce more reliable prediction results. Experiments on 5 challenging benchmark datasets reveal that the method outperforms current models in terms of performance, and runs at 38 fps, proving its effectiveness and efficiency.

Exploring the diversified teaching mode of vocal singing in colleges and universities by integrating audio-visual and multi-sensory senses

Abstract In this paper, starting from the fusion of audio-visual multisensory vocal music feature representation, the RAE algorithm is used to represent the music lyrics text in vocal singing teaching. Due to the heterogeneity of the feature space of the audio modality and the lyrics modality, which makes it exceptionally tricky to directly mine the correlation between these two modalities, it is necessary to optimize the research through the fusion of the audio and the textual modality for the representation of the implicit space of the music. According to the SVM’s optimal classification, hyperplane is not limited by the data dimension. Combined with the kernel function parameters and emotional LFSM fusion, a visual multisensory vocal singing teaching emotion model based on SVM is constructed. The parameter environment and dataset are selected, and the comparison method and evaluation criteria are determined to analyze emotional research on vocal singing teaching in colleges and universities. The results show that in terms of model performance, the SVM model in this paper is 8.5% higher than model 6, reaching the highest 0.873, with stronger emotion extraction and recognition ability, greatly improving the emotion classification results of the model. The multimedia type of audio singing material is the least helpful for expression and physical performance in vocal singing teaching, with a total value of 67. This study provides a more comprehensive understanding of the emotional changes of students in teaching activities so as to identify problems, improve and optimize them, and give more students guidance and support in performing vocal singing.

Analysis of the Practical Path of Vocational Education Serving Rural Revitalization by Digital Means and Philosophy of Technology

Abstract In this paper, based on digital means and technical philosophy, the histogram algorithm in the Light GBM model is used to calculate the floating point values of the raw data for the analysis of vocational education service for rural revitalization so that each of its features is converted into a histogram. In order to prevent the feature training overfitting problem of the Light GBM model, the LightGBM model is optimized by the iterative tree MPA algorithm, and the prediction model of returning to poverty risk based on IMPA-LightGBM is constructed. Starting from the current situation of vocational education service for rural revitalization, we put forward research hypotheses to realize the research design of vocational education accurate poverty alleviation service for rural revitalization and carry out an example analysis of vocational education service for rural revitalization combined with digital technology. The results show that in terms of model performance, the WAPE values of the return-to-poor risk values obtained from the prediction of the IMPA-LightGBM model are all lower than 5.5%, so the prediction effect is relatively satisfactory, and the return-to-poor risk values of poverty-eradicating households can be effectively predicted. On the practice road analysis, the standard deviation (SD) of the rural revitalization development index in China as a whole decreased from 0.63 to 0.52, which means that the differences in rural revitalization among provinces are decreasing. This study explores the synergistic development of the community of interest between vocational education and rural revitalization through the cultivation of new vocational farmers.

A COMPARISON OF PRE-PROCESSING APPROACHES FOR REMOTELY SENSED TIME SERIES CLASSIFICATION BASED ON FUNCTIONAL ANALYSIS

Abstract. Satellite remote sensing has gained a key role for vegetation mapping distribution. Given the availability of multi-temporal satellite data, seasonal variations in vegetation dynamics can be used trough time series analysis for vegetation distribution mapping. These types of data have a very high variability within them and are subjected by artifacts. Therefore, a pre-processing phase must be performed to properly detect outliers, for data smoothing process and to correctly interpolate the data. In this work, we compare four pre-processing approaches for functional analysis on 4-years of remotely sensed images, resulting in four time series datasets. The methodologies presented are the results of the combination of two outlier detection methods, namely tsclean and boxplot functions in R and two discrete data smoothing approaches (Generalized Additive Model ”GAM” on daily and aggregated data). The approaches proposed are: tsclean-GAM on aggregated data (M01), boxplot-GAM on aggregated data (M02), tsclean-GAM on daily data (M03), boxplot-GAM on daily data (M04). Our results prove that the approach which involves tsclean function and GAM applied to daily data (M03) is ameliorative to the logic of the procedure and leads to better model performance in terms of Overall Accuracy (OA) which is always among the highest when compared with the others obtained from the other three different approaches.

Analysis of vehicle pedestrian crash severity using advanced machine learning techniques

In 2015, over 17% of pedestrians were killed during vehicle crashes in Hong Kong while it raised to 18% from 2017 to 2019 and expected to be 25% in the upcoming decade. In Hong Kong, buses and the metro are used for 89% of trips, and walking has traditionally been the primary way to use public transportation. This susceptibility of pedestrians to road crashes conflicts with sustainable transportation objectives. Most studies on crash severity ignored the severity correlations between pedestrian-vehicle units engaged in the same impacts. The estimates of the factor effects will be skewed in models that do not consider these within-crash correlations. Pedestrians made up 17% of the 20,381 traffic fatalities in which 66% of the fatalities on the highways were pedestrians. The motivation of this study is to examine the elements that pedestrian injuries on highways and build on safety for these endangered users. A traditional statistical model's ability to handle misfits, missing or noisy data, and strict presumptions has been questioned. The reasons for pedestrian injuries are typically explained using these models. To overcome these constraints, this study used a sophisticated machine learning technique called a Bayesian neural network (BNN), which combines the benefits of neural networks and Bayesian theory. The best construction model out of several constructed models was finally selected. It was discovered that the BNN model outperformed other machine learning techniques like K-Nearest Neighbors, a conventional neural network (NN), and a random forest (RF) model in terms of performance and predictions. The study also discovered that the time and circumstances of the accident and meteorological features were critical and significantly enhanced model performance when incorporated as input. To minimize the number of pedestrian fatalities due to traffic accidents, this research anticipates employing machine learning (ML) techniques. Besides, this study sets the framework for applying machine learning techniques to reduce the number of pedestrian fatalities brought on by auto accidents.

Assessing industrial wastewater effluent toxicity using boosting algorithms in machine learning: A case study on ecotoxicity prediction and control strategy development

Trace heavy metals have a tendency to persist in the effluent of industrial wastewater treatment facilities, leading to toxic effects on downstream water bodies. Traditional assessment methods relied on animal testing, but ethical concerns have rendered them unacceptable. An alternative solution is to evaluate wastewater toxicity using trophic-level aquatic organisms as bioassays. However, these bioassay methods involve costly and time-consuming chemical and biological analytical experiments. In this study, an artificial intelligence-powered water quality assessment (AiWA) approach is proposed for predicting industrial effluent ecotoxicity to further enhance the quick and cost-effective ecotoxicity assessment process. Initially, 99 samples were collected from industrial wastewater treatment plants representing 21 different industries in the Republic of Korea. Fourteen parameters were measured, encompassing both physicochemical and ecotoxicological aspects. Boosting algorithms, especially extreme gradient boosting (XGBoost) and adaptive boosting (AdaBoost), were employed for model development. XGBoost outperformed AdaBoost in terms of model performance. Feature selection analysis revealed that conductivity, copper, lead, selenium, pH, and zinc concentrations were the most suitable inputs for training the boosting model. The innovated XGBoost-based AiWA model demonstrated significantly higher performance (i.e., up to 80%) compared to conventional models with an R2 value of exceeding 0.94 and root mean square error of 3.5 toxicity unit for predicting the integrated toxicity unit (ITU). Additionally, pH and conductivity emerged as crucial indicators for reflecting ecotoxicity levels. Specially, this case study indicated that non-toxic/directly dischargeable levels (TU ≤ 1) were achieved when the pH ranged from 6.8 to 8.4 and the conductivity remained below 1651 μS/cm. These findings are expected to facilitate rapid and cost-effective detection of heavy metal ecotoxicity in industrial wastewater effluents, aiding decision-making in wastewater management.

Modeling Edge Flame Propagation for Flame Particle Tracking in the Forced Ignition Process of Turbulent Nonpremixed Jet Flame

ABSTRACT To ensure safe and reliable operation, ignition performance in propulsion and power devices needs to be thoroughly assessed. Reduced order models (ROMs) for ignition have gained increasing attention, given that first-principles models are restricted by immense computational costs that preclude rapid assessment for the design purpose. Currently, flame self-propagation is generally neglected in the ROM descriptions of ignition process and the impact on the flame front dynamics as well as the ignition probability map is not clear for low swirl or jet flames which are of interest for industrial applications. In this study, an empirical model for flame self-propagation is proposed to augment the capability of Lagrangian flame particle tracking method, a ROM model to predict ignition probability based on nonreacting flow field. The model adopts the classic scaling law of turbulent flame speed together with fuel gradient correction and flame self-propagation to mimic the edge flame dynamics during the ignition process. The model performances in terms of flame front dynamics and ignition probability map are quantified in a turbulent nonpremixed methane-air jet flame. Results reveal that only with flame self-propagation, the flame particle tracking can well capture the transient evolution of the leading flame front as observed in experiments. It is found that the predicted ignition probability map agrees well with experimental data with high ignition probability around the vicinity of the stoichiometric region at the downstream axial distance of 6 to 40 times the jet diameter. This study illustrates the importance to account for flame self-propagation to correctly predict ignition probability in jet dominant nonpremixed flames.

Analyzing the Integration Interpretation of Civics and Family Poetry Culture in University Language Programs Based on Deep Learning Models

Abstract In this paper, firstly, we start from the imagery in the poem, excavate the resources of the ideology and politics in the university language poetry, and put forward the feasible path of the integration of the ideology and politics of the university language course and the poetry culture of the family training. The research on poetry culture in university language is conducted using a deep learning model. Based on the deep learning model, keywords in the poetry text are extracted and extended, taking into account the issue of word splitting and lexical annotation. The sequence coding of word vectors is needed so as to construct a poetry generation model based on RNNPG, and then the research and analysis on the concept of ideology and politics of the poetry culture in the university language are carried out. The results show that in terms of model performance, the RNNPG model is 1.17 overall, and the overall generation effect is more ideal compared to other single discriminator generation models. The acceptance level of sophomores is generally above 0.29, which is better than that of freshmen, and the teaching of ancient poems under the perspective of “curriculum politics” is especially necessary to improve their cultural taste and aesthetic interest. This study is conducive to the personalized development of students and the formation of correct concepts.