Prediction Performance Of Model Research Articles

Hybrid machine learning approach to prediction of the compressive and flexural strengths of UHPC and parametric analysis Shapley additive explanations algorithm

Ultra-high-performance concrete (UHPC) is a sustainable construction material; it can be applied as a substitute for cement concrete. Artificial intelligence methods have been used to evaluate concrete composites to reduce time and money in the construction industries. So, this study applied machine learning (ML) and hybrid ML approaches to predict the compressive and flexural strength of UHPC. A dataset of 626 compressive strength and 317 flexural strength data points was collected from the published research articles, where fourteen important variables were selected as input parameters for the analysis of hybrid ML and ML algorithms. This research used XGBoost, LightGBM, and hybrid XGBoost- LightGBM algorithms to predict UHPC materials. Grid search (GS) techniques were used to adjust model hyper-parameters in search of improved high accuracy and efficiency. ML and hybrid ML models were train, and the test stage utilized statistical assessments such as R-square, root mean square error (RMSE), mean absolute error (MAE), and coefficient of efficiency (CE). The results presented hybrid ML algorithm was superior to the XGBoost and LightGBM algorithms in terms of R-square and RMSE values for both compressive strength and flexural strength prediction. A hybrid ML model and two ML models showed CS considerable R-square values above 0.94 at the testing stages and just over 0.97 at the training phase. Hybrid ML model performance accuracy for CS prediction was R-square value found that almost 0.996 for training and 0.963 for testing phases. At the same time, the FS prediction result showed that the R-square value of the Hybrid ML model and two traditional ML models were founded at almost 0.95 for the training phase and around 0.81 for the testing phase. But among them, the hybrid XGB-LGB model prediction performance was high accuracy and lowest error for CS and FS of UHPC trained and its hyperparameters optimized. Additionally, the SHAP investigation reveals the impact and relationship of the input variables with the output variables. SHAP analysis outcome reveals that curing age and steel fiber content input parameter had the highest positive impact on compressive strength and flexural strength of UHPC.

Predicting treatment response in multicenter non-small cell lung cancer patients based on federated learning

BackgroundMulticenter non-small cell lung cancer (NSCLC) patient data is information-rich. However, its direct integration becomes exceptionally challenging due to constraints involving different healthcare organizations and regulations. Traditional centralized machine learning methods require centralizing these sensitive medical data for training, posing risks of patient privacy leakage and data security issues. In this context, federated learning (FL) has attracted much attention as a distributed machine learning framework. It effectively addresses this contradiction by preserving data locally, conducting local model training, and aggregating model parameters. This approach enables the utilization of multicenter data with maximum benefit while ensuring privacy safeguards. Based on pre-radiotherapy planning target volume images of NSCLC patients, a multicenter treatment response prediction model is designed by FL for predicting the probability of remission of NSCLC patients. This approach ensures medical data privacy, high prediction accuracy and computing efficiency, offering valuable insights for clinical decision-making.MethodsWe retrospectively collected CT images from 245 NSCLC patients undergoing chemotherapy and radiotherapy (CRT) in four Chinese hospitals. In a simulation environment, we compared the performance of the centralized deep learning (DL) model with that of the FL model using data from two sites. Additionally, due to the unavailability of data from one hospital, we established a real-world FL model using data from three sites. Assessments were conducted using measures such as accuracy, receiver operating characteristic curve, and confusion matrices.ResultsThe model’s prediction performance obtained using FL methods outperforms that of traditional centralized learning methods. In the comparative experiment, the DL model achieves an AUC of 0.718/0.695, while the FL model demonstrates an AUC of 0.725/0.689, with real-world FL model achieving an AUC of 0.698/0.672.ConclusionsWe demonstrate that the performance of a FL predictive model, developed by combining convolutional neural networks (CNNs) with data from multiple medical centers, is comparable to that of a traditional DL model obtained through centralized training. It can efficiently predict CRT treatment response in NSCLC patients while preserving privacy.

Development and Validation of a Predictive Model for Early Identification of Cognitive Impairment Risk in Community-Based Hypertensive Patients.

Objective: To investigate the risk factors for the development of mild cognitive dysfunction in hypertensive patients in the community and to develop a risk prediction model. Method: The data used in this study were obtained from two sources: the China Health and Retirement Longitudinal Study (CHARLS) and the Chinese Longitudinal Healthy Longevity Survey (CLHLS). A total of 1121 participants from CHARLS were randomly allocated into a training set and a validation set, following a 70:30 ratio. Meanwhile, an additional 4016 participants from CLHLS were employed for external validation of the model. The patients in this study were divided into two groups: those with mild cognitive impairment and those without. General information, employment status, pension, health insurance, and presence of depressive symptoms were compared between the two groups. LASSO regression analysis was employed to identify the most predictive variables for the model, utilizing 14-fold cross-validation. The risk prediction model for cognitive impairment in hypertensive populations was developed using generalized linear models. The model's discriminatory power was evaluated through the area under the receiver operating characteristic (ROC) curve and calibration curves. Results: In the modeling group, eight variables such as gender, age, residence, education, alcohol use, depression, employment status, and health insurance were ultimately selected from an initial pool of 21 potential predictors to construct the risk prediction model. The area under the curve (AUC) values for the training, internal, and external validation sets were 0.777, 0.785, and 0.782, respectively. All exceeded the threshold of 0.7, suggesting that the model effectively predicts the incidence of mild cognitive dysfunction in community-based hypertensive patients. A risk prediction model was developed using a generalized linear model in conjunction with Lasso regression. The model's performance was evaluated using the area under the receiver operating characteristic (ROC) curve. Hosmer-Lemeshow test values yielded p = .346 and p = .626, both of which exceeded the 0.05 threshold. Calibration curves demonstrated a significant agreement between the nomogram model and observed outcomes, serving as an effective tool for evaluating the model's predictive performance. Discussion: The predictive model developed in this study serves as a promising and efficient tool for evaluating cognitive impairment in hypertensive patients, aiding community healthcare workers in identifying at-risk populations.

Enabling active visitor management: local, short-term occupancy prediction at a touristic point of interest

AbstractAfter the temporary shock of the Covid-19 pandemic, the rapid recovery and resumed growth of the tourism sectors accelerates unsustainable tourism, resulting in local (over-)crowding, environmental damage, increased emissions, and diminished tourism acceptance. Addressing these challenges requires an active visitor management system at points of interest (POI), which requires local and timely POI-specific occupancy predictions to predict and mitigate crowding. Therefore, we present a new approach to measure visitor movement at an open-spaced, and freely accessible POI and evaluate the prediction performance of multiple occupancy and visitor count machine learning prediction models. We analyze multiple case combinations regarding spatial granularity, time granularity, and prediction time horizons. With an analysis of the SHAP values we determine the influence of the most important features on the prediction and extract transferable knowledge for similar regions lacking visitor movement data. The results underline that POI-specific prediction is achievable with a moderate relation for occupancy prediction and a strong relation for visitor count prediction. Across all cases, XGBoost and Random Forest outperform other models, with prediction accuracy increasing as the prediction time horizon shortens. For effective active visitor management, combining multiple models with different spatial aggregations and prediction time horizons provides the best information basis to identify appropriate steering measures. This innovative application of digital technologies facilitates information exchange between destination management organizations and tourists, promoting sustainable destination development and enhancing tourism experience.

Enhanced prediction of cement raw meal oxides by near-infrared spectroscopy using machine learning combined with chemometric techniques.

The component analysis of raw meal is critical to the quality of cement. In recent years, near-infrared (NIR) has been emerged as an innovative and efficient analytical method to determine the oxide content of cement raw meal. This study aims to utilize NIR spectroscopy combined with machine learning and chemometrics to improve the prediction of oxide content in cement raw meal. The Savitzky-Golay convolution smoothing method is applied to eliminate noise interference for the analysis of calcium carbonate ( ), silicon dioxide ( ), aluminum oxide ( ), and ferric oxide ( ) in cement raw materials. Different wavelength selection techniques are used to perform a comprehensive analysis of the model, comparing the performance of several wavelength selection techniques. The back-propagation neural network regression model based on particle swarm optimization algorithm was also applied to optimize the extracted and screened feature wavelengths, and the model prediction performance was checked and evaluated using and RMSE. In conclusion, the results indicate that NIR spectroscopy in combination with ML and chemometrics has great potential to effectively improve the prediction performance of oxide content in raw materials and highlight the importance of modeling and wavelength selection techniques. By enabling more accurate and efficient determination of oxide content in raw materials, NIR spectroscopy coupled with meta-modeling has the potential to revolutionize quality assurance practices in cement manufacturing.

Deep learning in the precise assessment of primary Sjögren's Syndrome based on ultrasound images.

This study aimed to investigate the value of a deep learning (DL) model based on greyscale ultrasound (US) images for precise assessment and accurate diagnosis of primary Sjögren's syndrome (pSS). This was a multicentre prospective analysis. All pSS patients were diagnosed according to 2016 ACR/EULAR criteria. 72 pSS patients and 72 sex- and age-matched healthy controls recruited between January 2022 and April 2023, together with 41 patients and 41 healthy controls recruited from June 2023 to February 2024 were used for DL model development and validation, respectively. DL model was constructed based on the ResNet 50, input with preprocessed all participants' bilateral submandibular glands (SMGs), parotid glands (PGs), and lacrimal glands (LGs) greyscale US images. Diagnostic performance of the model was compared with two radiologists. The accuracy of prediction and identification performance of DL model were evaluated by calibration curve. 864 and 164 greyscale US images of SMGs, PGs, and LGs were collected for development and validation of the model. The AUCs of DL model in the SMG, PG, and LG were 0.92, 0.93, 0.91 in the model cohort, and were 0.90, 0.88, 0.87 in the validation cohort respectively, outperforming both radiologists. Calibration curves showed the prediction probability of DL model were consistent with the actual probability in both model cohort and validation cohort. DL model based on greyscale US images showed diagnostic potential in the precise assessment of pSS patients in the SMG, PG, and LG, outperforming conventional radiologist evaluation.

ARFGCN: Adaptive Receptive Field Graph Convolutional Network for Urban Crowd Flow Prediction

Urban crowd flow prediction is an important task for transportation systems and public safety. While graph convolutional networks (GCNs) have been widely adopted for this task, existing GCN-based methods still face challenges. Firstly, they employ fixed receptive fields, failing to account for urban region heterogeneity where different functional zones interact distinctly with their surroundings. Secondly, they lack mechanisms to adaptively adjust spatial receptive fields based on temporal dynamics, which limits prediction performance. To address these limitations, we propose an Adaptive Receptive Field Graph Convolutional Network (ARFGCN) for urban crowd flow prediction. ARFGCN allows each region to independently determine its receptive field size, adaptively adjusted and learned in an end-to-end manner during training, enhancing model prediction performance. It comprises a time-aware adaptive receptive field (TARF) gating mechanism, a stacked 3DGCN, and a prediction layer. The TARF aims to leverage gating in neural networks to adapt receptive fields based on temporal dynamics, enabling the predictive network to adapt to urban regional heterogeneity. The TARF can be easily integrated into the stacked 3DGCN, enhancing the prediction. Experimental results demonstrate ARFGCN’s effectiveness compared to other methods.

Establishment of a risk prediction model for olfactory disorders in patients with transnasal pituitary tumors by machine learning

To construct a prediction model of olfactory dysfunction after transnasal sellar pituitary tumor resection based on machine learning algorithms. A cross-sectional study was conducted. From January to December 2022, 158 patients underwent transnasal sellar pituitary tumor resection in three tertiary hospitals in Sichuan Province were selected as the research objects. The olfactory status was evaluated one week after surgery. They were randomly divided into a training set and a test set according to the ratio of 8:2. The training set was used to construct the prediction model, and the test set was used to evaluate the effect of the model. Based on different machine learning algorithms, BP neural network, logistic regression, decision tree, support vector machine, random forest, LightGBM, XGBoost, and AdaBoost were established to construct olfactory dysfunction risk prediction models. The accuracy, precision, recall, F1 score, and area under the ROC curve (AUC) were used to evaluate the model's prediction performance, the optimal prediction model algorithm was selected, and the model was verified in the test set of patients. Of the 158 patients, 116 (73.42%) had postoperative olfactory dysfunction. After missing value processing and feature screening, an essential order of influencing factors of olfactory dysfunction was obtained. Among them, the duration of operation, gender, type of pituitary tumor, pituitary tumor apoplexy, nasal adhesion, age, cerebrospinal fluid leakage, blood scar formation, and smoking history became the risk factors of olfactory dysfunction, which were the key indicators of the construction of the model. Among them, the random forest model had the highest AUC of 0.846, and the accuracy, precision, recall, and F1 score were 0.750, 0.870, 0.947, and 0.833, respectively. Compared with the BP neural network, logistic regression, decision tree, support vector machine, LightGBM, XGBoost, and AdaBoost, the random forest model has more advantages in predicting olfactory dysfunction in patients after transnasal sellar pituitary tumor resection, which is helpful for early identification and intervention of high-risk clinical population, and has good clinical application prospects.

Analysis of Model Predictive Control-Based Energy Management System Performance to Enhance Energy Transmission

A supervisory control system using Model Predictive Control (MPC) has been designed to evaluate the efficiency of wind and solar power and is consistent with the cost function in the supervisory MPC optimization problem. A two-layer Economic Model Predictive Control (EMPC) framework has been developed and has improved results such as cost reductions compared to recent advanced methods. A speed Generalized Predictive Control (GPC) scheme intended for wind energy conversion systems was developed last year, with simulation results indicating superior performance over previous models. A Hierarchical Distributed Model Predictive Control (HDMPC) can work under different weather conditions with improved economic performance and keep a good balance between power delivery and load demand. An energy management system (EMS), built on the basis of MPC, can be quite lucrative for the sphere in the present climate scenario, with the selection and testing of suitable algorithms, controlled processes, cost functions, and a set of constraints as well as with proper optimizations carried out. Previous research indicates that an MPC-based EMS has the potential to be a good solution to manage energy well and also introduced it to the world experimentally. The key intention of this research study is to explore the existing advances that have been introduced and to analyze their performance in terms of cost function, different sets of constraints, variant conversion processes, and scalability to achieve more optimized operation of MPC-based EMS.

Prognostic factors and predictive model construction in patients with non-small cell lung cancer: a retrospective study

ObjectiveThe purpose of this study was to construct a nomogram model based on the general characteristics, histological features, pathological and immunohistochemical results, and inflammatory and nutritional indicators of patients so as to effectively predict the overall survival (OS) and progression-free survival (PFS) of patients with non-small cell lung cancer (NSCLC) after surgery.MethodsPatients with NSCLC who received surgical treatment in our hospital from January 2017 to June 2021 were selected as the study subjects. The predictors of OS and PFS were evaluated by univariate and multivariable Cox regression analysis using the Cox proportional risk model. Based on the results of multi-factor Cox proportional risk regression analysis, a nomogram model was established using the R survival package. The bootstrap method (repeated sampling for 1 000 times) was used to internally verify the nomogram model, and C-index was used to represent the prediction performance of the nomogram model. The calibration graph method was used to visually represent its prediction compliance, and decision curve analysis (DCA) was used to evaluate the application value of the model.ResultsUnivariate and multivariate analyses were used to identify independent prognostic factors and to construct a nomogram of postoperative survival and disease progression in operable NSCLC patients, with C-index values of 0.927 (907–0.947) and 0.944 (0.922–0.966), respectively. The results showed that the model had high predictive performance. Calibration curves for 1-year, 2-year, and 3-year OS and PFS show a high degree of agreement between the predicted probability and the actual observed probability. In addition, the results of the DCA curve show that the model has good clinical application value.ConclusionWe established a predictive model of survival prognosis and disease progression in patients with non-small cell lung cancer after surgery, which has good predictive performance and can guide clinicians to make the best clinical decision.

Federated Learning on Distributed and Encrypted Data for Smart Manufacturing

Abstract Industry 4.0 drives exponential growth in the amount of operational data collected in factories. These data are commonly distributed and stored in different business units or cooperative companies. Such data-rich environments increase the likelihood of cyber attacks, privacy breaches, and security violations. Also, this poses significant challenges on developing machine learning models on sensitive data that are distributed among different business units. To fill this gap, this paper presents a novel privacy-preserving framework to enable federated learning on siloed and encrypted data for smart manufacturing. Specifically, we leverage fully homomorphic encryption (FHE) to allow for computation on ciphertexts and generate encrypted results which, when decrypted, match the results of mathematical operations performed on the plaintexts. Multi-layer encryption and privacy protection reduce the likelihood of data breaches while maintaining the prediction performance of machine learning models. Experimental results in real-world case studies show that the proposed framework yields superior performance to reduce the risk of cyber attacks and harness siloed data for smart manufacturing.

Development and verification of a full physiologically-based pharmacokinetic model for sublingual buprenorphine in healthy adult volunteers that accounts for nonlinear bioavailability.

Sublingual buprenorphine is used for opioid use disorder and neonatal opioid withdrawal syndrome (NOWS). The study aimed to develop a full physiologically-based pharmacokinetic (PBPK) model that can adequately describe dose- and formulation-dependent bioavailability of buprenorphine. Simcyp (v21.0) was used for model construction. Four linear regression models (i.e. untransformed or log-transformed for dose or proportion sublingually absorbed) were explored to describe sublingual absorption of buprenorphine across dose. Published clinical trial data not used in model development were used for verification. The PBPK model's predictive performance was deemed adequate if the geometric means of ratios between predicted and observed (P/O ratios) area under the curve (AUC), peak concentration (Cmax), and time to reach Cmax (Tmax) fell within the 1.25-fold prediction error range. Sublingual buprenorphine absorption was best described by a regression model with logarithmically transformed dose. By integrating this nonlinear absorption profile, the PBPK model adequately predicted buprenorphine pharmacokinetics (PK) following administration of sublingual tablets and solution across a dose range of 2-32 mg, with geometric mean (95% confidence interval) P/O ratios for AUC and Cmax equaling 0.99 (0.86-1.12) and 1.24 (1.09-1.40), respectively, and median (5th to 95th percentile) for Tmax equaling 1.11 (0.69-1.57). A verified PBPK model was developed that adequately predicts dose- and formulation-dependent buprenorphine PK following sublingual administration. Significance Statement The PBPK model developed in this study is the first to adequately predict dose- and formulation-dependent sublingual buprenorphine pharmacokinetics. Accurate prediction was facilitated by the incorporation of a novel nonlinear absorption model. The developed model will serve as the foundation for fetomaternal PBPK modeling to predict maternal and fetal buprenorphine exposures to optimize buprenorphine treatment for neonatal opioid withdrawal syndrome (NOWS).

Modeling and predicting energy consumption of chiller based on dynamic spatial-temporal graph neural network

As an important equipment in the HVAC system, the chiller accounts for 50–60 % of the total energy consumption. According to the characteristics of chiller operation data and using the multivariate time series (MTS) forecasting method, a novel chiller energy consumption prediction model (T-CADGCN) is proposed, which is based on the temporal convolution (TC) and the dynamic graph convolution (DGC) network with a channel attention (CA) mechanism. Two datasets are used to verify the prediction performance of different models. Compared with the actual energy consumption, the average MAPEs of the T-CADGCN model on the measured dataset and simulated dataset are 0.92 % and 0.08 %, respectively. Whereas they are 2.8 % and 0.91 % for LSTM, 11.45 % and 11.74 % for MTGNN_GCA, 12.3 % and 12.63 % for MTGNN, respectively. The results demonstrate that the T-CADGCN model can effectively enhance the prediction performance of chiller energy consumption, followed by LSTM, then MTGNN_GCA, and finally MTGNN. Besides, the prediction performance on different future time steps is also conducted on two datasets. It can be found that all models decrease in prediction accuracy with the increase of the prediction time steps, but the T-CADGCN model can maintain a relatively good prediction performance with high robustness. The changes of in-degree and out-degree in the chiller energy consumption node in the dynamic graph construction module indicate that the factors affecting and being affected by chiller energy consumption change over time. This study can help to predict the energy consumption of chiller accurately and thus contributes to building energy management.

Response prediction and probabilistic analysis of the vehicle-ballasted track system considering track irregularity based on long-short term memory neural network

It is necessary to efficiently and accurately predict the dynamic behavior of the train-track system subjected to massive random track geometric excitations for system evaluation and maintenance. This paper proposed an engineering practical approach for the train-track system behavior prediction by integrating the deep learning surrogate model-based long-short term memory (LSTM) and physical model of the train-ballasted track interaction (TBTI) system, and then the generalized probability density evolution method (GPDEM) is introduced to achieve the probabilistic assessment of the TBTI system under limited track irregularities, where the track irregularity probability model (TIPM) coupled with generalized-F discrepancy points sets is employed to compress the track irregularity sets to reduce the number of system responses (as training target) but without loss of the responses completeness, therefore, it can promote the efficiency and accuracy of training model. The prediction performance of the surrogate model is illustrated by comparing the time-frequency domain information, mean and standard deviation of the responses obtained by the physical model and surrogate model. The results indicated the proposed method can accurately predict the system response, especially the low and medium frequency response below 50 Hz, and improve the efficiency of stochastic analysis for the TBTI system by approximately 20 times. Furthermore, the applicability of the proposed method is further examined by investigating the influence of probability levels of track irregularity and train speeds on train performance. Its extended model is also implemented to predict the response of systems induced by any measured track irregularity in real-time, which promotes the generalization capability of the proposed surrogate model.

Adaptive SARIMA modelling for continuous chamber temperature tracking in ultra-low temperature freezers

Ultra-low temperature (ULT) freezers are vital for storing perishable biological materials, requiring continuous monitoring of chamber thermal conditions to ensure sample integrity. A reliable dynamic data-driven model is key for smart proactive surveillance of the ULT freezers through real-time status tracking. However, the dynamic nature of ULT freezers, influenced by external disturbances, demands adaptive models to sustain performance. Complex models thus face challenges due to time-consuming parameter updates. To ensure model prediction performance while minimising model updating time, this study proposes a Seasonal Autoregressive Integrated Moving Average (SARIMA) model-based approach for short-term ULT freezer chamber temperature prediction. This method solely relies on historical chamber temperature data, eliminating the need for additional measurements. Moreover, two adaptive duty cycle length (DCL) selection strategies and a one-step ahead DCL estimation method are developed to address auto-correlation issues caused by recurring duty cycles. Comparative analyses with reference models demonstrate that the SARIMA model, incorporating future DCL effects, excels in one-step prediction and multi-step forecast under stable and variant DCL periods. Moreover, it outperforms the reference models in auto-correlation representation, affirming its reliability in predicting future states. The proposed modelling approach promises an effective method for freezing chamber temperature prediction, enabling continuous status tracking and early fault detection. Moreover, it holds great potential for modelling other critical temperatures in refrigeration systems and facilitating flexible energy operations.

Influence of usage and model inaccuracies on the performance of smart hot water heaters: lessons learned from a demand response field test

Domestic hot water heaters are considered to be easily integrated as flexible loads for demand response. While literature grows on reproducible simulation and lab tests, real-world implementation in field tests considering state estimation and demand prediction-based model predictive control approaches is rare. This work reports the findings of a field test with 16 autonomous smart domestic hot water heaters. The heaters were equipped with a retrofittable sensor/actuator setup and a real-time price-driven model predictive control unit, which covers state estimation, demand prediction, and optimization of switching times. With the introduction of generic performance indicators (specific costs and thermal efficiency), the results achieved in the field are compared by simulations to standard control modes (instantaneous heating, hysteresis, night-only switching). To evaluate how model predictive control performance depends on the user demand prediction and state estimation accuracy, simulations assuming perfect predictions and state estimations are conducted based on the data measured in the field. Results prove the feasible benefit of RTP-based model predictive control in the field compared to a hysteresis-based standard control regarding cost reduction and efficiency increase but show a strong dependency on the degree of utilization.

Feature engineering of environmental covariates improves plant genomic-enabled prediction.

Because Genomic selection (GS) is a predictive methodology, it needs to guarantee high-prediction accuracies for practical implementations. However, since many factors affect the prediction performance of this methodology, its practical implementation still needs to be improved in many breeding programs. For this reason, many strategies have been explored to improve the prediction performance of this methodology. When environmental covariates are incorporated as inputs in the genomic prediction models, this information only sometimes helps increase prediction performance. For this reason, this investigation explores the use of feature engineering on the environmental covariates to enhance the prediction performance of genomic prediction models. We found that across data sets, feature engineering helps reduce prediction error regarding only the inclusion of the environmental covariates without feature engineering by 761.625% across predictors. These results are very promising regarding the potential of feature engineering to enhance prediction accuracy. However, since a significant gain in prediction accuracy was observed in only some data sets, further research is required to guarantee a robust feature engineering strategy to incorporate the environmental covariates.

Compressive strength of waste-derived cementitious composites using machine learning

Abstract Marble cement (MC) is a new binding material for concrete, and the strength assessment of the resulting materials is the subject of this investigation. MC was tested in combination with rice husk ash (RHA) and fly ash (FA) to uncover its full potential. Machine learning (ML) algorithms can help with the formulation of better MC-based concrete. ML models that could predict the compressive strength (CS) of MC-based concrete that contained FA and RHA were built. Gene expression programming (GEP) and multi-expression programming (MEP) were used to build these models. Additionally, models were evaluated by calculating R 2 values, carrying out statistical tests, creating Taylor’s diagram, and comparing theoretical and experimental readings. When comparing the MEP and GEP models, MEP yielded a slightly better-fitted model and better prediction performance (R 2 = 0.96, mean absolute error = 0.646, root mean square error = 0.900, and Nash–Sutcliffe efficiency = 0.960). According to the sensitivity analysis, the prediction of CS was most affected by curing age and MC content, then by FA and RHA contents. Incorporating waste materials such as marble powder, RHA, and FA into building materials can help reduce environmental impacts and encourage sustainable development.

Advanced tree-based machine learning methods for predicting the seismic response of regular and irregular RC frames

In this study, predicting the maximum drift ratio (MDR) for two types of structures: 8-story regular (R8) and vertically irregular (IR8) reinforced concrete (RC) frames are focused. To accomplish this, advanced tree-based machine learning (ML) methods, namely Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and Stochastic Gradient Boosting (SGB) are utilized. 2300 ground motion (GM) records, includes twenty-one input parameters and their nonlinear time history analyses results are considered as a dataset. The performance of the RF, XGBoost, and SGB models was examined considering the impact of the data pre-processing issue to test whether the use of data pre-processing approaches in model development could improve the prediction performance of the advanced ML models. Various data pre-processing techniques, including outlier detection using boxplot, missing data imputation using RF, data transformation using log transformation, and feature selection using Hill Climbing are employed as part of ML pipeline. The models' performance is evaluated using six different well-established statistical evaluation metrics. Subsequently, the results are systematically compared to identify the most effective prediction model. Our findings highlight the substantial impact of data pre-processing on the performance of ML models when predicting seismic responses, specifically the MDR of RC frames. Notably, RF consistently outperformed other ML models for both R8 and IR8 structures in the case of the simple ML pipeline strategy and data pre-processing approach. The RF model achieved R2 scores of R2 = 0.87 and R2 = 0.91 on the test data before implementing data pre-processing for R8 and IR8, respectively. After data pre-processing, R2 scores showed an improvement of 12% and 6.22% in the cases of R8 and IR8. In contrast, the performance scores of SGB consistently fell behind those of RF and XGBoost models.

Multi Task Learning Based Shallow Parsing for Indian Languages

Shallow Parsing is an important step for many Natural Language Processing tasks. Although shallow parsing has a rich history for resource rich languages, it is not the case for most Indian languages. Shallow Parsing consists of POS Tagging and Chunking. Our study focuses on developing shallow parsers for Indian languages. As part of shallow parsing we included morph analysis as well. For the study, we first consolidated available shallow parsing corpora for 7 Indian Languages (Hindi, Kannada, Bangla, Malayalam, Marathi, Urdu, Telugu) for which treebanks are publicly available. We then trained models to achieve state of the art performance for shallow parsing in these languages for multiple domains. Since analyzing the performance of model predictions at sentence level is more realistic, we report the performance of these shallow parsers not only at the token level, but also at the sentence level. We also present machine learning techniques for multitask shallow parsing. Our experiments show that fine-tuned contextual embedding with multi-task learning improves the performance of multiple as well as individual shallow parsing tasks across different domains. We show the transfer learning capability of these models by creating shallow parsers (only with POS and Chunk) for Gujarati, Odia, and Punjabi for which no treebanks are available. As a part of this work, we will be releasing the Indian Languages Shallow Linguistic (ILSL) benchmarks for 10 Indian languages including both the major language families Indo-Aryan and Dravidian as common building blocks that can be used to evaluate and understand various linguistic phenomena found in Indian languages and how well newer approaches can tackle them.