adaptiveboosting (AdaBoost) algorithms fuse multiple weak classifiers to generate a strong classifier by adaptively determining the fusion weights of the weak classifiers. According to incorrect or correct classifica...
详细信息
adaptiveboosting (AdaBoost) algorithms fuse multiple weak classifiers to generate a strong classifier by adaptively determining the fusion weights of the weak classifiers. According to incorrect or correct classification results, sample weights become larger or smaller. However, this weight update scheme neglects valuable information in the results. Moreover, an important requirement for weak classifiers is an accuracy higher than random guessing. This requirement is likely to lead to an unexpected result. This means that several generated weak classifiers with similar classification results cannot learn from each other. Consequently, the advantage of fusing multiple weak classifiers disappears. The classification and therefore distinction of different failure modes in materials is a typical task for classical nondestructive testing approaches as well as for new approaches based on machine learning methods. In the case different approaches can be applied, the main question is, which and how tuned approaches provide the best results in terms of accuracy or similar metrics. In the multi-damage classification task distinguishing different physical failure mechanisms in Carbon Fiber Reinforced Polymer (CFRP), the above two aspects complicate the application of AdaBoost algorithms. To improve the results, a novel AdaBoost with distance-based weighted least square support vector machine (WLSSVM) and filter factor is proposed. The distance-based WLSSVM is employed to increase the diversity of weak classifiers, the distances of the classification plane and samples are used to measure the classification task. The filter factor is proposed to filter out unnecessary classifiers contributing less to the final decision. The experimental results demonstrate that the improved AdaBoost schemes with distance-based WLSSVM and filter factor outperform state-of-the-art algorithms. The effects of the scheme using the new weighted update and the filter factor on the algorithm are discus
Tracing technology is increasingly being used in fluvial and aeolian sediment provenance assessments. Using synthetic sample mixtures in validations of unmixing model performance is becoming a standard step in sedimen...
详细信息
Tracing technology is increasingly being used in fluvial and aeolian sediment provenance assessments. Using synthetic sample mixtures in validations of unmixing model performance is becoming a standard step in sediment source fingerprinting. With tracer variability fully considered in the sources and target sediment, this study explored a semiempirical modelling strategy based on virtual sediment mixtures and the adaptiveboosting (AdaBoost) algorithm. The obtained integrated unmixing model contained multiple composite fingerprints, and the weight coefficient was obtained from the iterative process. The performance of the integrated unmixing model was compared with that of unmixing models applying single or equal-weighted multiple composite fingerprints. All generated virtual sediment mixtures (280) were split into a training dataset (240) and a test dataset (40) to validate the generalization ability of the models. The results showed that the integrated unmixing model achieved better performance than the unmixing models with a single composite fingerprint (basic models). The integrated unmixing model yielded an average mean absolute error (A-MAE) of 5.51% for training data and 5.72% for test data, and it achieved better accuracy than the best basic model (6.23% on the training and 6.75% on the test dataset, composite fingerprint C-1) or equal-weighted model (6.32% on the training and 6.41% on the test dataset, average of 42 basic models). The robustness of the model accuracy was also improved with the AdaBoost algorithm. The modelling approach proposed in this study has the potential to maximize the use of all tracer information and further improve the reliability of sediment fingerprinting.
Group key management offers a flexible and reliable security mechanism for secure communication in wireless sensor network by assisting with suitable adjustments of the number of keys per node and the number of re-key...
详细信息
Group key management offers a flexible and reliable security mechanism for secure communication in wireless sensor network by assisting with suitable adjustments of the number of keys per node and the number of re-keying messages. In this article, we obtained a datasets using a projective plane after removing a single element. We employ a stacking ensemble algorithm to predict the re-keying value in a projective plane. To improve the performance of the prediction in the stacking model, adaptiveboosting and random forest models are chosen as base learners, and for the meta-learner, linear regression is chosen. We observed that the stacking ensemble algorithm demonstrated higher accuracy compared to individual models. The accuracy of the stacking ensemble algorithm is found to be 0.9999, with MAE, MSE, and RMSE values of 0.0026, 0.0000, and 0.0030 respectively.
For the measurement of gas volume fraction in natural gas wells, a strategy based on the fusion of arrayed fiberoptic probes (AFOP) and artificial intelligence algorithms is proposed to enhance the precision and effic...
详细信息
For the measurement of gas volume fraction in natural gas wells, a strategy based on the fusion of arrayed fiberoptic probes (AFOP) and artificial intelligence algorithms is proposed to enhance the precision and efficiency of gas volume fraction monitoring. As a key front-end component for obtaining gas phase information, AFOP determines the optimal structure by analyzing its performance metrics in bubble capture and its interference with fluid flow. A back-end gas volume fraction prediction model was constructed using a machine learning algorithm. The model first uses a particle swarm optimization (PSO) algorithm to enhance the backpropagation (BP) neural network as a weak predictor and then integrates multiple weak predictors through the adaptiveboosting (AdaBoost) algorithm to create a strong predictor. The experimental results show that compared with the support vector machine (SVM), BP neural network, and PSO-BP neural network, the PSO-BP-AdaBoost algorithm has advantages in prediction precision, with a maximum relative error of only 0.14 %, providing a more effective solution for research and application in related fields.
Predicting floor water inrush has become increasingly challenging as coal is being mined at greater depths. We established a practical predictive method using a hybrid artificial intelligence (AI) model and geographic...
详细信息
Predicting floor water inrush has become increasingly challenging as coal is being mined at greater depths. We established a practical predictive method using a hybrid artificial intelligence (AI) model and geographic information system (GIS) techniques. The hybrid model is a classifier that mixes a back propagation neural network (BPNN) with an adaptive boosting algorithm (AdaBoost). To assess the effectiveness of the model, 33 borehole data points with known water inrush results in the Yangcheng coal mine were used as data samples to train and test the model. The outcomes demonstrated a predictive accuracy of 100%, far exceeding the accuracy and stability of the BPNN classifier alone using the same parameters. Then, GIS techniques were used to extend the approach throughout the mining region;the greatest risk was shown to be in the middle of the area. Given the limited data set, errors may exist in extending the risk predictions for the entire mining area, so more data needs to be collected to ensure the accuracy of subsequent predictions. Still, we believe that the methods and steps adopted in this study can be used to create practical predictive models in different mining regions.
Depressive symptoms are common in mild cognitive impairment (MCI), dementia caused by Alzheimer's disease (AD dementia) and in cognitively unimpaired older adults. However, it is unclear whether they could contrib...
详细信息
Depressive symptoms are common in mild cognitive impairment (MCI), dementia caused by Alzheimer's disease (AD dementia) and in cognitively unimpaired older adults. However, it is unclear whether they could contribute to the identification of cognitive impairment in ageing. To assess the potential utility of depressive symptoms to distinguish between healthy cognitive ageing and MCI and AD dementia. The diagnostic workup of the cognitive function of 1737 older cognitively unimpaired individuals, 334 people with MCI and 142 individuals with AD dementia relied on a comprehensive neuropsychiatric assessment, including the Mini Mental State Examination (MMSE). Depressive symptoms were tapped with the 15-item Geriatric Depression Scale (GDS). Proportional odds logistic regression (POLR) models and the machine learning technique adaptive boosting algorithm (AdaBoost) were employed. Stratified repeated random subsampling (stratified bootstrap resampling) was used to recursive partitioning to training- and validation set (70/30 ratio). The average accuracy of the POLR models for the GDS total score in distinguishing between cognitive impairment and healthy cognitive ageing exceeded 78% and was inferior to that of MMSE. Of note, the sensitivity of GDS total score was very low. By employing the AdaBoost algorithm and considering GDS items separately, the average accuracy was higher than 0.72 and comparable to that of the MMSE, while sensitivity- and specificity values were more balanced. The findings of the study provide initial evidence that depressive symptoms may contribute to distinguishing between cognitive impairment and cognitively healthy ageing.
We propose an efficient boostingalgorithm for multiclass classification, called ***, that extends SAMME [14] and AdaBoost [4]. The algorithm iteratively applies the weak learnability condition of SAMME to eliminate c...
详细信息
ISBN:
(纸本)9798400704901
We propose an efficient boostingalgorithm for multiclass classification, called ***, that extends SAMME [14] and AdaBoost [4]. The algorithm iteratively applies the weak learnability condition of SAMME to eliminate classes to find the correct classificiation. The iterative weak learnability is a sufficient and necessary condition for boostability, but it is also easier to validate than the EOR criterion of *** [9]. We show that the training error of *** vanishes at the exponential rate, while the generalization error converges to zero at the same rate as AdaBoost. *** numerically outperforms SAMME and achieves performance comparable to *** on benchmark datasets.
Short-term power load forecasting plays a vital role in the planning of distribution network and the development of social economy. Many researchers have devoted their attention to construct point forecasting models. ...
详细信息
Short-term power load forecasting plays a vital role in the planning of distribution network and the development of social economy. Many researchers have devoted their attention to construct point forecasting models. However, the traditional point forecasting regards the forecasting result as a deterministic variable. Since the deviation existed in load forecasting is simply unavoidable and significant, which has great volatility and randomness, point forecast methods are difficult to capture the fluctuation of power load. Probability forecast models are proposed to obtain the uncertain information in the load power. In this paper, firstly, some error information is obtained from the deterministic forecasting results of point forecasting;secondly, the interval of time series data is divided according to the deterministic error information;finally, the Bootstrap method is used to estimate the confidence interval of the deteministic error information to obtain the uncertainty information in the power load data. The instability and randomness of the deterministic error are otained by combining the interval forecasting method so as to improve the accuracy of power load forecasting. Therefore, probability forecasting is combined with point forecasting to obtain more accurate results. The proposed model is used to forecast power load for Queensland, Australia and capital district of New York State. The experimental results show that the proposed method performs better than other comparative models. The experimental results show that the ELM-AdaBoost model has better forecast performance in both long-term and short-term load datasets, and can overcome the seasonality of power load time series data. (C) 2022 The Author(s). Published by Elsevier Ltd.
Recent developments in high-performance concrete (HPC) have made it a high-tech material with enhanced durability and properties, and it is a more ecological material with a longer life cycle. Conventional approaches ...
详细信息
Recent developments in high-performance concrete (HPC) have made it a high-tech material with enhanced durability and properties, and it is a more ecological material with a longer life cycle. Conventional approaches to assessing HPC properties and environmental effects may lack precision in predicting the effects of additional cementitious materials. adaptive boosting algorithm (ADA) and Stochastic Gradient boosting (SGB) have emerged as promising approaches for predicting HPC properties and environmental effects, offering more accurate and efficient alternative methods. In this study, the ADA and SBG models have been employed to predict the two essential HPC mechanical properties, including compressive and tensile strengths (CS and TS) and Global Warming Potential (GWP) by coupling with meta-heuristic algorithms, namely Stadium Spectators Optimizer (SSO), Golden Jackal algorithm (GJA), Brown-bear Optimization algorithm (BOA), and Golf Optimization algorithm (GOA)(sic) for developing hybrid and ensemble prediction approaches. The study assessed the effectiveness of different models for predicting HPC properties and GWP by employing various metrics. The SGSBGG framework showcased remarkable precision and performance compared to other models in the prediction of CS, TS, and GWP. It outshone its counterparts by achieving the highest prediction accuracy, securing R-2 values of 0.994, 0.994, and 0.995 for CS, TS, and GWP predictions, respectively. Additionally, it demonstrated unparalleled accuracy with the lowest modeling errors, registering an RMSE of 1.306 for CS, 0.097 for TS, and 4.469 for GWP. Despite the clear superiority of the SGSBGG model, it is noteworthy that the SGSO, as a hybrid model, delivered notably compatible results.
Coalbed methane (CBM) productive efficiency and coal mine disasters such as gas outbursts and water inrush are closely correlated with coal seam permeability. Effective prediction of coal seam permeability can provide...
详细信息
Coalbed methane (CBM) productive efficiency and coal mine disasters such as gas outbursts and water inrush are closely correlated with coal seam permeability. Effective prediction of coal seam permeability can provide guidance for CBM production and prevention of coal mine disasters. In this research, a hybrid neural network prediction model integrating a genetic algorithm, an adaptive boosting algorithm, and a back propagation neural network was developed to predict coal seam permeability. Additional momentum and variable learning rate algorithms were used to improve the learning rate and accuracy of the model, and the model structure was optimized, including the number of hidden layer nodes and the transfer function. The input parameters of the prediction model included gas pressure, compressive strength, reservoir temperature, and effective stress. The corresponding output parameter was coal seam permeability. The correlation between the parameters was calculated. Additionally, a comparative analysis between the proposed prediction model and four other prediction models was carried out to demonstrate the advantages of the proposed model. The results indicated that the correlations between compressive strength, gas pressure, reserve temperature, effective stress, and coal seam permeability were 0.334, -0.148, -0.406, and -0.785, respectively. The proposed prediction model had high accuracy compared with the other prediction models, and its coefficient of determination and root mean squared error were 0.999 and 0.021. Thus, the model can predict coal seam permeability more accurately.
暂无评论