Grid Search Research Articles

Comparison of time-to-event machine learning models in predicting biliary complication and mortality rate in liver transplant patients

Post-Liver transplantation (LT) survival rates stagnate, with biliary complications (BC) as a major cause of death. We analyzed longitudinal data with a median 19-month follow-up. BC was diagnosed with ultrasounds and MRCP. Missing data was imputed using mean and median. Data preprocessing involved feature scaling and one-hot encoding. Survival analysis used filter (Cox-P, Cox-c) and embedded (RSF, LASSO) feature selection methods. Seven survival machine learning algorithms were used: LASSO, Ridge, RSF, E-NET, GBS, C-GBS, and FS-SVM. Model development employed 5-fold cross-validation, random oversampling, and hyperparameter tuning. Random oversampling addressed data imbalance. Optimal hyperparameters were determined based on average C-index. Features importance was assessed using standardized regression coefficients and permutation importance for top models. Stability was evaluated using 5-fold cross-validation standard deviation. Finally, 1799 observations with 40 outcome predictors were included. RSF with Ridge achieved the highest performance (C-index: 0.699) for BC prediction, while RSF with RSF had the highest performance (C-index: 0.784) for mortality prediction. Top BC predictors were LT graft types, IBD in recipients, recipient’s BMI, recipient’s history of PVT, and previous LT history. For mortality, they were post-transplant AST, creatinine, recipient’s age, post-transplant ALT, and tacrolimus consumption. We identified BC and mortality risk factors, improving decision-making and outcomes.

Machine Learning: A Multicentre Study on Predicting Lateral Lymph Node Metastasis in cN0 Papillary Thyroid Carcinoma.

The necessity of prophylactic lateral neck dissection for cN0 papillary thyroid carcinoma (PTC) remains debated. This study aimed to compare traditional nomograms with machine learning (ML) models for predicting ipsilateral lateral and level II, III, and IV lymph node metastasis (LNM). Data from 1616 PTC patients diagnosed via fine needle aspiration biopsy from Hospital A were split into training and testing sets (7:3). 243 patients from Hospital B served as validation set. Four dependent variables-ipsilateral lateral, level II, III, and IV LNM-were analyzed. Eight ML models (Logistic Regression, Decision Tree, Random Forest-RF, Gradient Boosting, Support Vector Machine, K-Nearest Neighbor, Gaussian Naive Bayes, Neural Networks) were developed and validated using 10-fold cross-validation and grid search hyperparameter tuning. Models were assessed using 11 metrics including accuracy, area under the curve (AUC), specificity, and sensitivity. The best was compared with nomograms using the Probability-based Ranking Model Approach (PMRA). RF outperformed other approaches achieving accuracy, AUC, specificity, and sensitivity of 0.773/0.728, 0.858/0.799, 0.984/0.935, 0.757/0.807 in the testing/validation sets respectively for ipsilateral LLNM. A streamlined model based on the top ten contributing features that includes ipsilateral central lymph node metastasis rate, extrathyroidal extension, and ipsilateral central lymph node metastasis number retained strong performance and clearly surpassed a traditional nomogram approach based on multiple metrics and PMRA analysis. Similar results were obtained for the other dependent variables, with the RF models relying on distinct but overlapping sets of features. Clinical tool implementation is facilitated via a web-based calculator for each of the four dependent variables. ML, especially RF, reliably predicts lateral LNM in cN0 PTC patients, outperforming traditional nomograms.

Metaheuristic-enhanced Deep Learning Model for Accurate Alzheimer's Disease Diagnosis from MRI Imaging

Alzheimer’s Disease (AD) is the neuro-degenerative dementia, where the precise and early recognition of AD is vital for timely treatment to reduce mortality rate. A new automated model is implemented in this work for early discovery of AD in the Magnetic Resonance Imaging (MRI) brain scans. Initially, the input brain scans are taken from the Alzheimer's disease Neuroimaging Initiative (ADNI) database. Further, the acquired raw brain scans are visually improved by employing the binary normalization technique. The denoised brain scans are fed to the pre-trained Convolutional Neural Network (CNN) named GoogleNet for feature extraction. Next, the extracted richer feature values are fed to the Long Short Term Memory (LSTM) network for classifying the brain scan as Normal Control (NC), Mild Cognitive Impairment (MCI) and AD. In this manuscript, a Honey Badger Optimization Algorithm (HBOA) technique is incorporated with the LSTM networks for hyper-parameters optimization, where this procedure helps in diminishing the LSTM network’s complexity and computational time. The experimental results conducted on the ADNI database underscore the HBOA-based LSTM network's effectiveness, showcasing a remarkable mean classification accuracy of 97.83% in multi-class classification. Moreover, the sensitivity of HBOA based LSTM for AD/NC is 96.73% which is high when compared to the existing methodologies such as SVM with radial basis kernel function and NCSINs. This performance surpasses that of other comparative models for AD detection, emphasizing the superior capabilities and potential of the proposed method in the early detection.

Prediction of traffic time using XGBoost model with hyperparameter optimization

Prediction of traffic time using XGBoost model with hyperparameter optimization

A Novel Monthly Runoff Prediction Model Based on KVMD and KTCN-LSTM-SA

Due to the uncertainty of meteorological factors and the influence of human activities, the monthly runoff series often exhibit the characteristics of non-stationarity. The appropriate prediction model and the hyperparameters of the model are often difficult to determine, and this affects the model prediction performance. For obtaining the accurate runoff prediction results, a novel prediction model (KVMD-KTCN-LSTM-SA) is proposed. This hybrid model uses Kepler optimization algorithm (KOA)-optimized Variable Mode Decomposition (KVMD), KOA-optimized temporal convolutional network–long short-term memory (TCN-LSTM), and the self-attention (SA) mechanism. KVMD effectively reduces the difficulty of predicting the monthly runoff series, KOA helps to find the optimal hyperparameters of the model, TCN is combined with LSTM, and the SA mechanism effectively increases the performance of the model. Monthly runoff from three hydrological stations in the Hetian River basin and one hydrological station in the Huaihe River basin are predicted with the proposed model, and six models are selected for comparison. The KVMD-KTCN-LSTM-SA model effectively reduces runoff fluctuation and combines the advantages of multiple models and achieves satisfactory runoff prediction results. During the testing period, the proposed model achieves NSE of 0.978 and R2 of 0.982 at Wuluwati station, NSE of 0.975 and R2 of 0.986 at Tongguziluoke station, and NSE of 0.978 and R2 of 0.982 at Jiangjiaji station. The proposed hybrid model provides a new approach for monthly runoff prediction, which is capable of better managing and predicting mid-long-term runoff.

Performance-informed urban design: training a generalized surrogate model for predicting building energy demand across residential morphologies in Singapore

Building performance simulations (BPS) are essential for managing rising energy demands but are computationally intensive, limiting their use in early-stage design. Surrogate models (SMs) offer fast approximations using machine learning but lack generalizability for changing building morphologies. This study introduces three morphology-based feature extraction methods – Zernike image moments, polar raycasting, and principal component analysis – to train a generalizable SM for predicting annual energy use intensity (EUI) of 2.5D residential buildings in Singapore. The best SM achieved a 9.2% RMSPE, with features yielding low error and avoiding overfitting. This framework enables efficient design space exploration, aiding architects and urban designers in optimizing masterplans for energy efficiency. In the final section, we demonstrate a design space exploration (DSE) framework with our generalizable SM using a novel parametric model, showcasing its speed, adaptability and convenience over conventional optimization workflows. Highlights Novel approach explores building morphology's impact to annual EUI . Using Zernike moments, ray casting and PCA to extract features from any 2.5D buildings. 4000 tuned SMs trained from hyperparameter optimisation. Most robust SM achieved 9.2% RMSPE when tested with 800 samples. SM's generalizability is proven by applying it to a design model with previously unseen parameters.

CART-ANOVA-Based Transfer Learning Approach for Seven Distinct Tumor Classification Schemes with Generalization Capability

Background/Objectives: Deep transfer learning, leveraging convolutional neural networks (CNNs), has become a pivotal tool for brain tumor detection. However, key challenges include optimizing hyperparameter selection and enhancing the generalization capabilities of models. This study introduces a novel CART-ANOVA (Cartesian-ANOVA) hyperparameter tuning framework, which differs from traditional optimization methods by systematically integrating statistical significance testing (ANOVA) with the Cartesian product of hyperparameter values. This approach ensures robust and precise parameter tuning by evaluating the interaction effects between hyperparameters, such as batch size and learning rate, rather than relying solely on grid or random search. Additionally, it implements seven distinct classification schemes for brain tumors, aimed at improving diagnostic accuracy and robustness. Methods: The proposed framework employs a ResNet18-based knowledge transfer learning (KTL) model trained on a primary dataset, with 20% allocated for testing. Hyperparameters were optimized using CART-ANOVA analysis, and statistical validation ensured robust parameter selection. The model’s generalization and robustness were evaluated on an independent second dataset. Performance metrics, including precision, accuracy, sensitivity, and F1 score, were compared against other pre-trained CNN models. Results: The framework achieved exceptional testing accuracy of 99.65% for four-class classification and 98.05% for seven-class classification on the source 1 dataset. It also maintained high generalization capabilities, achieving accuracies of 98.77% and 96.77% on the source 2 datasets for the same tasks. The incorporation of seven distinct classification schemes further enhanced variability and diagnostic capability, surpassing the performance of other pre-trained models. Conclusions: The CART-ANOVA hyperparameter tuning framework, combined with a ResNet18-based KTL approach, significantly improves brain tumor classification accuracy, robustness, and generalization. These advancements demonstrate strong potential for enhancing diagnostic precision and informing effective treatment strategies, contributing to advancements in medical imaging and AI-driven healthcare solutions.

Effect of blade inclination angle for straight-bladed vertical-axis wind turbines

Abstract. Vertical-axis wind turbines (VAWTs) have received renewed research interest in the offshore environment due to a number of design synergies that have the potential to decrease the cost of energy for offshore wind. Many studies have been completed on the rotor design for straight-bladed rotors (H-rotors); however, there is sparse information on the effect of blade inclination angle on VAWT aerodynamic performance and the optimal blade design of VAWTs with inclined blades (V-rotors) for maximum power capture. This paper presents a systematic study into the effect of blade inclination angle and chord distribution on VAWT performance for different aspect ratios using a 3D implementation of the 2D actuator cylinder model which has been previously validated against higher-fidelity methods. A systematic approach based on a grid search is used to allow the wider design space to be studied and trends identified. In the case of fixed-chord-length blades, it is found that significant power gains are available through blade inclination, between 12 % and 71 % depending on blade length. This is driven by the increase in rotor swept area. Further investigation indicates that despite this, under maximum blade stress limitations the most economical solution for fixed-chord-length blades is H-rotors. Optimal chord distributions, which maximise the rotor power coefficient, are then obtained, and a natural blade taper is observed. For rotors with optimal chord distributions, similar power gains are observed through blade inclination, again between 12 % and 71 % depending on blade length. However, rotor configurations with the largest power gains are found to have significantly increased blade mass. For a given power rating, whilst satisfying limitations on maximum blade root bending stress, it is found that blade volume, a proxy for the blade mass, can be reduced between 16 % and 42 % dependent on blade length, and rotor torque can be reduced between 4 % and 9 %. This indicates the potential of V-rotors to reduce the cost of energy compared to H-rotors in traditional VAWT designs. Additionally, inclined blades are shown to increase the operational tip-speed ratio, demonstrating their applicability to turbines using secondary rotors, such as the X-rotor, where high tip-speed ratios are required for efficient power conversion between primary and secondary rotors.

PhotoD with LSST: Stellar Photometric Distances Out to the Edge of the Galaxy

Abstract As demonstrated with the Sloan Digital Sky Survey (SDSS), Pan-STARRS, and most recently with Gaia data, broadband near-UV to near-IR stellar photometry can be used to estimate distance, metallicity, and interstellar dust extinction along the line of sight for stars in the Galaxy. Anticipating photometric catalogs with tens of billions of stars from Rubin's Legacy Survey of Space and Time (LSST), we present a Bayesian model and pipeline that build on previous work and can handle LSST-sized datasets. Likelihood computations utilize MIST/Dartmouth isochrones and priors are derived from TRILEGAL-based simulated LSST catalogs from P. Dal Tio et al. The computation speed is about 10 ms per star on a single core for both optimized grid search and Markov Chain Monte Carlo methods; we show in a companion paper by K. Mrakovčić et al. how to utilize neural networks to accelerate this performance by up to an order of magnitude. We validate our pipeline, named PhotoD (in analogy with photo-z, photometric redshifts of galaxies) using both simulated catalogs and SDSS, DECam, and Gaia photometry. We intend to make LSST-based value-added PhotoD catalogs publicly available via the Rubin Science Platform with every LSST data release.

Optimizing high-strength concrete compressive strength with explainable machine learning

This study leverages machine learning to enhance the prediction of high-strength concrete (HSC) compressive strength, addressing the limitations of conventional methods, which are often tedious, less reliable, and time-consuming. Extreme Gradient Boosting (XGB) serves as the primary model, with hyperparameter optimization via metaheuristic algorithms such as Cuckoo Search (CSA), Water Strider (WS), Leopard Seal (LS), Harris Hawk (HH), Invasive Weed (IW), and Forest Optimization (FO). A total of 681 data sets were collected from existing literature. The models underwent tenfold cross-validation, with the LS-XGB model achieving an almost ideal performance in testing sets. Other models, including CSA-XGB, WS-XGB, HH-XGB, IW-XGB, and FO-XGB, also demonstrated strong performance, each with R2 > 0.96. For model explainability, Shapley's Additive Explanation (SHAP) analysis has been applied to the best-performing LS-XGB model. The analysis revealed that cement and superplasticizer (SP) are the most crucial features contributing to HSC development, with optimal ranges identified at 600–900 kg/m3 for cement and 8–10 kg/m3 for SP. The study demonstrates on how feature interactions contribute to concrete materials compressive strength, providing better and above all sustainable constructions. Furthermore, the LS-XGB model's optimal performance depicts the strongly nonlinear nature of HSC materials, validated through a set of derived graphs. Additionally, 30 concrete cubes were prepared for experimental validation, and the datasets demonstrated an accuracy of 92% showcasing the ability of models to make well informed decision.

Unlocking the link: predicting cardiovascular disease risk with a focus on airflow obstruction using machine learning

BackgroundRespiratory diseases and Cardiovascular Diseases (CVD) often coexist, with airflow obstruction (AO) severity closely linked to CVD incidence and mortality. As both conditions rise, early identification and intervention in risk populations are crucial. However, current CVD risk models inadequately consider AO as an independent risk factor. Therefore, developing an accurate risk prediction model can help identify and intervene early.MethodsThis study used the National Health and Nutrition Examination Survey (NHANES) III (1988–1994) and NHANES 2007–2012 datasets. Inclusion criteria were participants aged over 40 with complete AO and CVD data; exclusions were those with missing key data. Analysis included 12 variables: age, gender, race, PIR, education, smoking, alcohol, BMI, hyperlipidemia, hypertension, diabetes, and AO. Logistic regression analyzed the association between AO and CVD, with sensitivity and subgroup analyses. Six ML models predicted CVD risk for the general population, using AO as a predictor. RandomizedSearchCV with 5-fold cross-validation was used for hyperparameter optimization. Models were evaluated by AUC, accuracy, precision, recall, F1 score, and Brier score, with the SHapley Additive exPlanations (SHAP) enhancing explainability. A separate ML model was built for the subpopulation with AO, evaluated similarly.ResultsThe cross-sectional analysis showed that there was a significant positive correlation between AO occurrence and CVD prevalence, indicating that AO is an important risk factor for CVD (all P < 0.05). For the general population, the XGBoost model was selected as the optimal model for predicting CVD risk (AUC = 0.7508, AP = 0.3186). The top three features in terms of importance were age, hypertension, and PIR. For the subpopulation with airflow obstruction, the XGBoost model was also selected as the optimal model for predicting CVD risk (AUC = 0.6645, AP = 0.3545). SHAP shows that education level has the greatest impact on predicting CVD risk, followed by gender and race.ConclusionAO correlates positively with CVD. Age, hypertension, PIR affect CVD risk most in general. For AO patients, education, gender, ethnicity are key CVD risk factors.

Development of clinical decision support for patients older than 65 years with fall-related TBI using artificial intelligence modeling.

Older persons comprise most traumatic brain injury (TBI)-related hospitalizations and deaths and are particularly susceptible to fall-induced TBIs. The combination of increased frailty and susceptibility to clinical decline creates a significant ongoing challenge in the management of geriatric TBI. As the population ages and co-existing medical conditions complexify, so does the need to improve the quality of care for this population. Utilizing early hospital admission variables, this study will create and validate a multinomial decision tree that predicts the discharge disposition of older patients with fall-related TBI. From the National Trauma Data Bank, we retrospectively analyzed 11,977 older patients with a fall-related TBI (2017-2021). Clinical variables included Glasgow Coma Scale (GCS) score, intracranial pressure monitor use, venous thromboembolism (VTE) prophylaxis, and initial vital signs. Outcomes included hospital discharge disposition re-categorized into home, care facility, or deceased. Data were split into two sets, where 80% developed a decision tree, and 20% tested predictive performance. We employed a conditional inference tree algorithm with bootstrap (B = 100) and grid search options to grow the decision tree and measure discrimination ability using the area under the curve (AUC) and calibration plots. Our decision tree used seven admission variables to predict the discharge disposition of older TBI patients. Significant non-modifiable variables included total GCS and injury severity scores, while VTE prophylaxis type was the most important interventional variable. Patients who did not receive VTE prophylaxis treatment had a higher probability of death. The predictive performance of the tree in terms of AUC value (95% confidence intervals) in the training cohort for death, care, and home were 0.66 (0.65-0.67), 0.75 (0.73-0.76), and 0.77 (0.76-0.79), respectively. In the test cohort, the values were 0.64 (0.62-0.67), 0.75 (0.72-0.77), and 0.77 (0.73-0.79). We have developed and internally validated a multinomial decision tree to predict the discharge destination of older patients with TBI. This tree could serve as a decision support tool for caregivers to manage older patients better and inform decision-making. However, the tree must be externally validated using prospective data to ascertain its predictive and clinical importance.

A Machine Learning Approach to Volume Tracking in Multiphase Flow Simulations

This work presents a machine learning (ML) approach to volume-tracking for computational simulations of multiphase flow. It is an alternative to a commonly used procedure in the volume-of-fluid (VOF) method for volume tracking, in which the phase interfaces are reconstructed for flux calculation followed by volume advection. Bypassing the computationally expensive steps of interface reconstruction and flux calculation, the proposed ML approach performs volume advection in a single step, directly predicting the volume fractions at the next time step. The proposed ML function is two-dimensional and has eleven inputs. It was trained using MATLAB’s (R2021a) Deep Learning Toolbox with a grid search method to find an optimal neural network configuration. The performance of the ML function is assessed using canonical test cases, including translation, rotation, and vortex tests. The errors in the volume fraction fields obtained by the ML function are compared with those of the VOF method. In ideal conditions, the ML function speeds up the computations four times compared to the VOF method. However, in terms of overall robustness and accuracy, the VOF method remains superior. This study demonstrates the potential of applying ML methods to multiphase flow simulations while highlighting areas for further improvement.

Rethinking density ratio estimation based hyper-parameter optimization.

Hyper-parameter optimization (HPO) aims to improve the performance of machine learning algorithms by identifying appropriate hyper-parameters. By converting the computation of expected improvement into density-ratio estimation problems, existing works use binary classifiers to estimate these ratio and determine the next point by maximizing the class posterior probabilities. However, these methods tend to treat different points equally and ignore some important regions, because binary classifiers are unable to capture more information about search spaces and highlight important regions. In this work, we propose a hyper-parameter optimization method by estimating ratios and selecting the next point using multi-class classifiers. First, we divide all samples into multiple classes and train multi-class classifiers. The decision boundaries of the trained classifiers allow for a finer partitioning of search spaces, offering richer insights into the distribution of hyper-parameters within search spaces. We then define an acquisition function as a weighted sum of multi-class classifiers' outputs, with these weights determined by samples in each class. By assigning different weights to each class posterior probability in our acquisition function, points within search spaces are no longer treated equally. Experimental results on three representative tasks demonstrate that our method achieves a significant improvement in immediate regrets and convergence speed.

Dự báo giá cổ phiếu CATL đóng cửa và các chiến lược giao dịch sử dụng XGBoost và thuật toán tối ưu hóa Optuna

This study employs the XGBoost algorithm, enhanced through hyperparameter optimization with Optuna, to forecast the closing stock prices of CATL and develop effective trading strategies. The XGBoost model achieved superior performance, with an R² score of 0.9201, a Mean Absolute Error (MAE) of 0.1982, and a Mean Squared Error (MSE) of 0.0825, outperforming benchmark models such as ElasticNet and Decision Tree. The resulting trading strategy generated actionable buy, sell, and hold recommendations over a 30-day horizon, aiding investors in profit optimization. The findings highlight the high prediction accuracy and stability of the XGBoost model, particularly when optimized with Optuna, thereby enhancing decision-making efficiency in stock trading. Nonetheless, the study underscores the potential for further improvement through the integration of advanced deep learning models, alternative optimization techniques, or ensemble learning approaches. Future research could also explore incorporating unstructured data to refine forecasting accuracy and expand the applicability of trading strategies.

Accurate non-invasive detection of MASH with fibrosis F2-F3 using a lightweight machine learning model with minimal clinical and metabolomic variables

BackgroundThere are no known non-invasive tests (NITs) designed for accurately detecting metabolic dysfunction-associated steatohepatitis (MASH) with liver fibrosis stages F2-F3, excluding cirrhosis—the FDA-defined range for prescribing Resmetirom and other drugs in clinical trials. We aimed to validate and re-optimize known NITs, and most importantly to develop new machine learning (ML)-based NITs to accurately detect MASH F2-F3. MethodsClinical and metabolomic data were collected from 443 patients across three countries and two clinic types (metabolic surgery, gastroenterology/hepatology) covering the entire spectrum of biopsy-proven MASLD, including cirrhosis and healthy controls. Three novel types of ML models were developed using a categorical gradient boosting machine pipeline. These were compared with 24 biomarker, imaging, and algorithm-based NITs with both known cutoffs for MASH F2-F4 and re-optimized cutoffs for MASH F2-F3. ResultsNFS at a − 1.455 cutoff attained an AUC of 0.59, the highest sensitivity (90.9 %, 95 % CI 84.3–95.4), and NPV of 87.2 %. FIB4 risk stratification followed by elastography (8 kPa) had the best specificity (86.9 %) and PPV (63.3 %), with an AUC of 0.57. NFS followed by elastography improved the PPV to 65.3 % and AUC to 0.62. Re-optimized FibroScan-AST (FAST) at a 0.22 cutoff had the highest PPV (69.1 %). ML models using aminotransferases, metabolic syndrome components, BMI, and 3-ureidopropionate achieved an AUC of 0.89, which further increased to 0.91 following hyperparameter optimization and the addition of alpha-ketoglutarate. These new ML models outperformed all other NITs and displayed accuracy, sensitivity, specificity, PPV, and NPV up to 91.2 %, 85.3 %, 97.0 %, 92.4 %, and 90.7 % respectively. The models were reproduced and validated in a secondary sensitivity analysis, that used one of the cohorts as feature selection/training, and the rest as independent validation, likewise outperforming all other NITs. ConclusionsWe report for the first time the diagnostic characteristics of non-invasive, metabolomics-based biomarker models to detect MASH with fibrosis F2-F3 required for Resmetirom treatment and inclusion in ongoing phase-III trials. These models may be used alone or in combination with other NITs to accurately determine treatment eligibility.

Bayesian optimization enhanced FKNN model for Parkinson’s diagnosis

A progressive neurodegenerative condition that adversely impacts motor skills, speech, and cognitive abilities is Parkinson’s disease (PD). Research has revealed that verbal impediments manifest in the early of PD, making them a potential diagnostic marker. This study introduces an innovative approach, leveraging Bayesian Optimization (BO) to optimize a fuzzy k-nearest neighbor (FKNN) model, enhancing the detection of PD. BO-FKNN was validated on a speech datasets. To comprehensively evaluate the efficacy of the proposed model, BO-FKNN was compared against five commonly used parameter optimization methods, including FKNN based on Particle Swarm Optimization, FKNN based on Genetic algorithm, FKNN based on Bat algorithm, FKNN based on Artificial Bee Colony algorithm, and FKNN based on Grid search. Moreover, to further boost the diagnostic accuracy, a hybrid feature selection method based on Pearson Correlation Coefficient (PCC) and Information Gain (IG) was employed prior to the BO-FKNN method, consequently the PCCIG-BO-FKNN was proposed. The experimental outcomes highlight the superior performance of the proposed system, boasting an impressive classification accuracy of 98.47%.

AI-based plastic waste sorting method utilizing object detection models for enhanced classification.

The export ban on plastic waste by China has brought domestic plastic recycling to the forefront of environmental concerns, with sorting being a crucial step in the recycling process. This study assessed the performance of advanced AI models, Mask R-CNN, and YOLO v8, in enhancing plastic waste sorting. The models were evaluated in terms of accuracy, mean average precision (mAP), precision, recall, F1 score, and inference time, with hyperparameter tuning performed through grid search. Mask R-CNN, with an accuracy of 0.912 and mAP of 0.911, outperformed YOLO v8 in tasks requiring detailed segmentation, despite a longer inference time of 200-350ms. Conversely, YOLO v8, with an accuracy of 0.867 and mAP of 0.922, excelled in real-time applications owing to its shorter inference time of 80-160ms. This study underscores the importance of selecting the appropriate model based on specific application requirements.

The empirical analysis of multi-objective hyperparameter optimization in software vulnerability prediction

The security of software has always been a vital concern for software developers. Infringements on software systems can result in significant losses in terms of time and confidential data. Software vulnerabilities are considered to be the gateway for attackers to harm information systems. Hence, it is crucial to build effective software vulnerability prediction models. Machine learning algorithms produce high-performance prediction models but the performance of these models is affected by the hyperparameter settings of machine learning methods and imbalanced datasets. The current study aims to find out the effect of single- and multi-objective hyperparameter optimization on software vulnerability prediction models. The paper has proposed an experimental methodology that considers both optimizations and uses eight machine learning methods on PHP open-source public datasets (Drupal, Moodle, and PHPMyAdmin). The experimental results show that after applying multi-objective hyperparameter optimization, the highest AUC achieved is 0.9774 and the F1-Score attained is 0.9222, which is better than the benchmark studies. Random Forest has performed satisfactorily for all three datasets. The comparative analysis of single and multi-objective HPO is performed using post hoc Tukey’s HSD test. Furthermore, the effectiveness of the resampling technique ‘SMOTE’ is observed.

Hyperparameter optimization of convolutional neural network using particle swarm optimization for emotion recognition

Hyperparameter optimization of convolutional neural network using particle swarm optimization for emotion recognition