Noise Detection Method Research Articles

APPN: An Attention-based Pseudo-label Propagation Network for few-shot learning with noisy labels

Few-shot learning has garnered significant attention in deep learning as an effective approach for addressing the issue of data scarcity. Conventionally, training datasets in few-shot learning are clean. However, due to the occurrence of sensor malfunctions, data transmission anomalies, or inaccuracies in manual annotation, the accuracy of sample annotations cannot be guaranteed. Therefore, dealing with few-shot learning with noisy label becomes an urgent problem. To solve the above problem, we propose an Attention-based Pseudo-label Propagation Network (APPN). We make some technical contributions: (1) We propose an attention-based feature extraction method that can effectively capture the distinctions between clean and noisy samples. (2) We propose an improved graph-based pseudo-label propagation method that utilizes pseudo-labels with latent class information as initial labels, thereby enhancing the accuracy of label propagation. (3) We describe a comprehensive multi-step noise detection method which can accurately detect noisy samples from clean samples and effectively distinguish between out-of-domain (OOD) noise and in-domain (ID) noise. (4) Finally, the solid experiments verify the better performance on FC100, CIFAR-FS, miniImageNet, and tieredImageNet datasets. The results show that our method has high robustness and superiority in the few-shot learning with noisy labels. Code and models at: https://github.com/Typistchen/APPN.

Deep learning with noisy labels in medical prediction problems: a scoping review.

Medical research faces substantial challenges from noisy labels attributed to factors like inter-expert variability and machine-extracted labels. Despite this, the adoption of label noise management remains limited, and label noise is largely ignored. To this end, there is a critical need to conduct a scoping review focusing on the problem space. This scoping review aims to comprehensively review label noise management in deep learning-based medical prediction problems, which includes label noise detection, label noise handling, and evaluation. Research involving label uncertainty is also included. Our scoping review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We searched 4 databases, including PubMed, IEEE Xplore, Google Scholar, and Semantic Scholar. Our search terms include "noisy label AND medical/healthcare/clinical," "uncertainty AND medical/healthcare/clinical," and "noise AND medical/healthcare/clinical." A total of 60 papers met inclusion criteria between 2016 and 2023. A series of practical questions in medical research are investigated. These include the sources of label noise, the impact of label noise, the detection of label noise, label noise handling techniques, and their evaluation. Categorization of both label noise detection methods and handling techniques are provided. From a methodological perspective, we observe that the medical community has been up to date with the broader deep-learning community, given that most techniques have been evaluated on medical data. We recommend considering label noise as a standard element in medical research, even if it is not dedicated to handling noisy labels. Initial experiments can start with easy-to-implement methods, such as noise-robust loss functions, weighting, and curriculum learning.

Identifying Plausible Labels from Noisy Training Data for a Land Use and Land Cover Classification Application in Amazônia Legal

Most studies in the field of land use and land cover (LULC) classification in remote sensing rely on supervised classification, which requires a substantial amount of accurate label data. However, reliable data are often not immediately available, and are obtained through time-consuming manual labor. One potential solution to this problem is the use of already available classification maps, which may not be the true ground truth and may contain noise from multiple possible sources. This is also true for the classification maps of the MapBiomas project, which provides land use and land cover (LULC) maps on a yearly basis, classifying the Amazon basin into more than 24 classes based on the Landsat data. In this study, we utilize the Sentinel-2 data with a higher spatial resolution in conjunction with the MapBiomas maps to evaluate a proposed noise removal method and to improve classification results. We introduce a novel noise detection method that relies on identifying anchor points in feature space through clustering with self-organizing maps (SOM). The pixel label is relabeled using nearest neighbor rules, or can be removed if it is unknown. A challenge in this approach is the quantification of noise in such a real-world dataset. To overcome this problem, highly reliable validation sets were manually created for quantitative performance assessment. The results demonstrate a significant increase in overall accuracy compared to MapBiomas labels, from 79.85% to 89.65%. Additionally, we trained the L2HNet using both MapBiomas labels and the filtered labels from our approach. The overall accuracy for this model reached 93.75% with the filtered labels, compared to the baseline of 74.31%. This highlights the significance of noise detection and filtering in remote sensing, and emphasizes the need for further research in this area.

Robustness of Deep Learning models in electrocardiogram noise detection and classification

Background and Objective: Automatic electrocardiogram (ECG) signal analysis for heart disease detection has gained significant attention due to busy lifestyles. However, ECG signals are susceptible to noise, which adversely affects the performance of ECG signal analysers. Traditional blind filtering methods use predefined noise frequency and filter order, but they alter ECG biomarkers. Several Deep Learning-based ECG noise detection and classification methods exist, but no study compares recurrent neural network (RNN) and convolutional neural network (CNN) architectures and their complexity. Methods:This paper introduces a knowledge-based ECG filtering system using Deep Learning to classify ECG noise types and compare popular computer vision model architectures in a practical Internet of Medical Things (IoMT) framework. Experimental results demonstrate that the CNN-based ECG noise classifier outperforms the RNN-based model in terms of performance and training time. Results:The study shows that AlexNet, visual geometry group (VGG), and residual network (ResNet) achieved over 70% accuracy, specificity, sensitivity, and F1 score across six datasets. VGG and ResNet performances were comparable, but VGG was more complex than ResNet, with only a 4.57% less F1 score. Conclusions:This paper introduces a Deep Learning (DL) based ECG noise classifier for a knowledge-driven ECG filtering system, offering selective filtering to reduce signal distortion. Evaluation of various CNN and RNN-based models reveals VGG and Resnet outperform. Further, the VGG model is superior in terms of performance. But Resnet performs comparably to VGG with less model complexity.

Research on a synchronised classification method for loose particle detection signals of aerospace-sealed electronic components

Aerospace-sealed electronic components exhibit favourable anti-interference capability and high reliability and are widely utilised in satellites, rockets, and missiles. Loose particle detection is crucial to ensure high reliability. However, the classification problem of loose particle detection signals based on the particle impact noise detection (PIND) method has been a challenge for the high reliability of aerospace-sealed electronic components. To address this issue, this paper systematically proposes a synchronised classification method based on PIND and validates its feasibility. The proposed approach combines a self-developed empirical model (referred to as algorithm 1) with a classical machine learning model (referred to as algorithm 2). The key findings are as follows: this method has a recognition rate for loose particle signals of 91.86%. Compared to a single algorithmic mode, the overall speed improves by 300%. The maximal recognition accuracy rates for component signals, mixed signals, and excessive signals are 90.03%, 81.04%, and 95.17%, respectively. The test results demonstrate that the method effectively balances the accuracy and speed of loose particle detection and leverages the complementary advantages of the two algorithms, thereby addressing the multi-classification issue of loose particle detection signals.

Application of deep learning in noise detection of electromechanical products

In recent years, machine learning (ML) techniques have been widely integrated into manufacturing processes. This study explores supervised learning-based machine learning models designed for noise detection. The approach involves feature engineering, extracting relevant features from signals to provide meaningful inputs for machine learning models. Various algorithms, including convolutional neural networks (CNN) and other deep learning architectures, are employed to capture complex dependencies within the data. Training processes involve optimization techniques to enhance model performance and generalizability. Performance evaluation on benchmark datasets compares the proposed model with existing noise detection methods. Results demonstrate the outstanding accuracy and robustness of the generated machine learning model in distinguishing noise from clear signals. Additionally, the study explores the transferability of these models to new scenarios, highlighting their adaptability in practical applications. Findings from this research contribute to advancing noise detection methods, showcasing the effectiveness of machine learning in addressing this crucial challenge. The proposed solution is cost-effective, structurally simple, and holds vast potential for widespread implementation in factory settings in the future.

Signal detection and material identification method for loose particles inside aerospace relays based on overlapping signals

Loose particle is an important factor affecting the reliability of aerospace relays. Particle Impact Noise Detection method is a commonly used loose particle detection method, but it can generate interference signals. Accurately identifying loose particle signals and component signals becomes the key to accurately detecting loose particles. Meanwhile, deeply exploring the properties of loose particle signals and further provide material information can provide guidance for manufacturing processes. However, the existing loose particle signal and component signal identification research, as well as the loose particle material identification research, has problems such as limited research objects to pure signals, independent identification results, and failure to refer to the detection requirements in real application scenarios. It is difficult to apply and its practicality is not high. Based on this, the authors proposed a signal detection and material identification method for loose particles based on overlapping signals. Specifically, referring to the latest loose particle detection information, aerospace relay samples were made, and the component identification model and material identification model based on parameter-optimized random forest were trained, respectively. They can achieve good classification effects on data sets created from pure loose particle signals, pure component signals, and pure loose particle signals corresponding to different materials, respectively. On this basis, confidence coefficient was proposed to quantify the degradation degree of the classification effects of the two models on the data set created from overlapping signals, thus the component confidence coefficient and the loose particle confidence coefficient were obtained, respectively. They can be used to determine valid pulses from pure loose particle signals, pure component signals, and mixed signals in overlapping signals, completing loose particle detection. Valid pulses from pure loose particle signals were screened for material identification. In this way, for the aerospace relay to be tested, first, the loose particle detection results can be obtained by a comprehensive judgment of the component identification model, component confidence coefficient, and loose particle confidence coefficient. Second, the material identification results can be obtained by combining the material identification model with the majority voting processing. In addition, the definition of identification accuracy applicable to the loose particle detection was proposed to meet the engineering application requirements in real application scenarios. Multiple experiments verified the feasibility, practicality, and stability of the proposed method.

Variational Autoencoders for Biomedical Signal Morphology Clustering and Noise Detection.

Accurate estimation of physiological biomarkers using raw waveform data from non-invasive wearable devices requires extensive data preprocessing. An automatic noise detection method in time-series data would offer significant utility for various domains. As data labeling is onerous, having a minimally supervised abnormality detection method for input data, as well as an estimation of the severity of the signal corruptness, is essential. We propose a model-free, time-series biomedical waveform noise detection framework using a Variational Autoencoder coupled with Gaussian Mixture Models, which can detect a range of waveform abnormalities without annotation, providing a confidence metric for each segment. Our technique operates on biomedical signals that exhibit periodicity of heart activities. This framework can be applied to any machine learning or deep learning model as an initial signal validator component. Moreover, the confidence score generated by the proposed framework can be incorporated into different models' optimization to construct confidence-aware modeling. We conduct experiments using dynamic time warping (DTW) distance of segments to validated cardiac cycle morphology. The result confirms that our approach removes noisy cardiac cycles and the remaining signals, classified as clean, exhibit a 59.92% reduction in the standard deviation of DTW distances. Using a dataset of bio-impedance data of 97885 cardiac cycles, we further demonstrate a significant improvement in the downstream task of cuffless blood pressure estimation, with an average reduction of 2.67 mmHg root mean square error (RMSE) of Diastolic Blood pressure and 2.13 mmHg RMSE of systolic blood pressure, with increases of average Pearson correlation of 0.28 and 0.08, with a statistically significant improvement of signal-to-noise ratio respectively in the presence of different synthetic noise sources. This enables burden-free validation of wearable sensor data for downstream biomedical applications.

Abnormal noise identification of engines based on wavelet packet transform and bispectrum analysis

Abnormal noise is the most prominent problem for motorcycles and affects the consumers’ purchasing desire and driving experience and the enterprise’s competitiveness. Usually, the noise from a newly assembled engine is detected by manual auscultation (MA) to determine if the engine is operating normally. However, MA is also affected by subjective and objective factors with severe labor intensity, and its accuracy greatly fluctuates. Importantly, MA cannot be applied in a corporation with mass production and high-quality requirements. To improve the efficiency and accuracy of motorcycle engine quality inspection and achieve intelligent production, an online engine abnormal noise detection method was proposed based on wavelet packet transform (WPT) and bispectrum analysis (BA); this method improved the accuracy and stability of the identification of the abnormal noise engine and reduced the cost of the check. First, the acoustic signal of the engine of the motorcycle was acquired by using a free-field microphone. Second, the background noise of signals was eliminated by using the wavelet correlation coefficient (WCC) theory, and the signal features were extracted by applying WPT and BA. Third, the feature vectors were normalized before being used as support vector machine (SVM) samples. Fourth, an appropriate kernel function and parameters were selected to train the vector machine using the training sets. Finally, the testing sets were used to inspect the accuracy of the vector machines. The result showed that the training accuracy is 95% and the testing accuracy is 97.5 of the samples were suitable by using the method of wavelet packet transform-bispectrum analysis-support vector machines (WPT-BA-SVM). WPT-BA-SVM effectively identified engine fault types and provided the theoretical foundation for the establishment of an engine abnormal noise online detection system.

Uncertainty determination and reduction through novel approach for industrial IOT

The most fundamental and crucial processes in Data analysts is effective information processing while cleaning information from noisy sensors. Conventional outlier identification techniques should not be used directly for this type of assessment, as sensor data can contain both noise (piezoelectric microphone type) and severe abnormalities. To manage the problem of sound reduction while maintaining the anomalies in the IIoT information, this paper proposes a method of calculating the noise score which takes into account both the rate of variation and the variance. To establish the unit of evaluation of the measurement of difference which would be used in conjunction with statistical analyses, a sliding window method. Extensive testing shows that the proposed strategy surpasses existing advanced noise detection methods, producing a clean dataset that maintains anomalies to successfully use abnormal investigation techniques.

Research on Body Wave Extraction Technology of Mixed Active Source Ambient Noise

In complex surface environments, active source seismic exploration methods are difficult to stimulate and costly, while ambient noise detection technology is fast and convenient, with a little environmental damage and low cost. Therefore, it has aroused extensive research by scholars at home and abroad. At present, the use of passive source surface wave method for subsurface medium imaging has become mature. However, due to the low content and weak energy of the body wave signal in the actual ambient noise, it is particularly difficult to extract body waves using seismic interferometry. At the same time, body wave exploration has larger exploration depth and higher resolution. In this paper, we propose an ambient noise detection technology of mixed active sources, namely, imposing artificial seismic sources in the continuous ambient noise data acquisition process. We refer to the ambient noise of mixed active sources as "mixed ambient noise". It is found that the virtual refraction wave obtained by the ambient noise detection method of mixed active sources has high signal-to-noise ratio, strong anti-interference ability, and can obtain the real velocity information of the deeper subsurface medium, but cannot reflect the accurate time information. Supplementing the active source signal in the far field can obtain virtual refractions with higher signal-to-noise ratio, while effectively reducing the mixing phenomenon of surface wave and surface interference waves in the virtual common-shot record.

Snapping shrimp noise detection methods based on linear prediction analysis

This paper proposes features for detecting snapping shrimp noise based on linear predictive analysis. Snapping shrimps are a species inhabiting the ocean depths and are one of the main sources of underwater noise due to their high-amplitude signals that occur frequently. The proposed features utilize the characteristic of sudden onset and rapid decay of pistol shrimp noise by using linear predictive analysis to accurately detect the noise segments and mitigate the impact of snapping shrimp noise. The large error between the predicted values by linear predictive analysis and the actual measured values allows for effective detection of pistol shrimp segments. Additionally, the proposed features are combined with a fixed-threshold detector to further improve the performance of noise segment detection. When compared to the state-of-the-art method known as multilayer wavelet packet decomposition, the proposed method outperformed with an average improvement of 0.12 in terms of receiver operating characteristic curve and area under the curve. Moreover, the proposed method had lower computational complexity.

Fundamental study for vehicle abnormal sound detection and the direction estimation methods using machine learning with noise reduction

Detecting abnormal sounds during vehicle running condition is important to find out the vehicle’s problems at the early stage. In the mass production process, the detection test is generally carried out by the inspector. However, continuous inspection by a human for long time is hard and miss false inspection may occur. In addition, normal running noise, such as road noise, deteriorates the inspection accuracy. Therefore, an automatic accurate abnormal sound detection method under the normal noise is necessary to improve the detection accuracy with small man-hour. Furthermore, if the method has an ability to localize the abnormal sound direction, the countermeasure also becomes easy. In this study, we considered the noise detection method using machine learning with a noise reduction method and attempted the noise direction using two microphones. As a fundamental study, we applied these methods in an open room and the result showed that the method could detect the abnormal sound under normal noise condition and indicate the direction well in the simple verification test.

Probabilistic Matrix Factorization Recommendation Approach for Integrating Multiple Information Sources

Most previous studies on matrix factorization (MF)-based collaborative filtering (CF) have focused solely on user rating information for predicting recommendations. However, to further enhance the performance of recommender systems (RSs), it is important to also consider review information and rating reliability in the model. This article proposes a new probabilistic MF (PMF)-based CF method that integrates multiple information sources to provide reliable predictions. First, we introduce a sentiment-based PMF to handle user reviews and fit the normalized sentiment information obtained from our previously proposed sentiment analysis method. We also consider the helpfulness of reviews to highlight their reliability and effectiveness in this model. Subsequently, our proposed noise detection method is adopted to determine the reliability of user ratings, and then the rating matrix is transformed into a binary reliability matrix. A rating reliability-based PMF through Bernoulli distribution is then proposed to factorize it. To effectively integrate three types of information (ratings, reviews, and rating reliability) into a PMF procedure, we design a weight matrix using the proposed weighting strategy to generate a set of comprehensive prediction ratings with corresponding reliability probabilities. Experiments on four Amazon datasets demonstrate that our model outperforms comparison methods in terms of comprehensive evaluation.

Seismic random noise attenuation using DnCNN with stratigraphic dip constraint

Abstract In this paper, a method for seismic random noise detection and suppression using a denoising convolutional neural network (DnCNN) is presented. Thanks to the residual learning and batch normalization, deep learning networks can converge faster, the gradient descent and disappearance due to the increase of network layers are solved, and the residual results can be predicted more accurately. For seismic data, the variance estimation method is useful for obtaining an accurate noise distribution model and statistical parameters that provide a useful assessment of the noise level. With the variance estimation method based on weak texture blocks, a noise distribution model and statistical parameters can be derived with high accuracy, and this method effectively estimates seismic noise levels. The DnCNN networks are trained, and non-Gaussian noise reduction technology is used to achieve blind noise reduction at unknown levels, improving the noise reduction of seismic data. In addition, stratigraphic dip characteristics related to layer structure are used as DnCNN training network constraints to prevent effective signals in seismic data from being corrupted by conventional DnCNN noise reduction methods. Geological features such as faults and fracture-cavities can be effectively protected. Carbonate faults in the Tarim Basin in China are affected by the desert surface and the depth at which reservoirs are buried. The seismic data has a low signal-to-noise ratio, and the effective signals of the reservoir are low resolution. The seismic data can be effectively enhanced with this method for noise reduction in this area, the fracture-cavity is effectively displayed, and the fault features are also highlighted.

BAND: BAgging noise detectors with application to semiconductor wafer denoising

In semiconductor manufacturing processes, spatial defect patterns on semiconductor wafers can provide useful information to quality engineers regarding the root causes of abnormalities. As early detection of process problems increases the wafer yield, the automatic recognition of defect patterns is crucial. It its intrinsic assumption that a decluttering of noise defects helps in accurately classifying defect patterns. However, most existing noise detection methods require the pre-determination of parameters that significantly affect the detection performance. In this paper, we propose a parameter-free detection framework called BAND, which refers to BAgging Noise Detectors, to solve this problem. By adopting the framework of bootstrap aggregating (bagging), which combines the detection results from multiple outlier detection algorithms, the proposed framework ensures the diversity of randomly chosen parameter values while maintaining an accurate detection of noise defects. Based on the estimated distributions of outlierness scores, a group of outliers is discriminated from the other group of inliers via spectral clustering. This procedure enables the proposed framework to be parameter-free. Moreover, numerical experiments based on synthetic and real-world datasets are presented to demonstrate the effectiveness of the proposed framework.

Recognition Method for Electronic Component Signals Based on LR-SMOTE and Improved Random Forest Algorithm

<div>Loose particles are a major problem affecting the performance and safety of aerospace electronic components. The current particle impact noise detection (PIND) method used in these components suffers from two main issues: data collection imbalance and unstable machine-learning-based recognition models that lead to redundant signal misclassification and reduced detection accuracy. To address these issues, we propose a signal identification method using the limited random synthetic minority oversampling technique (LR-SMOTE) for unbalanced data processing and an optimized random forest (RF) algorithm to detect loose particles. LR-SMOTE expands the generation space beyond the original SMOTE oversampling algorithm, generating more representative data for underrepresented classes. We then use an RF optimization algorithm based on the correlation measure to identify loose particle signals in balanced data. Our experimental results demonstrate that the LR-SMOTE algorithm has a better data balancing effect than SMOTE, and our optimized RF algorithm achieves an accuracy of over 96% for identifying loose particle signals. The proposed method can also be popularized in the field of loose particle detection for large-scale sealing equipment and other various areas of fault diagnosis based on sound signals.</div>

Multi sensor moving image fusion analysis algorithm on the basis of neural network technology

Image fusion can extract the useful information of each channel about the same target to the maximum extent, and get high quality image. However, in this process, the image quality may be affected by noise and reduced. To reduce the image noise’s influence on the image fusion effect as well as improve the image fusion quality, a multi sensor moving image fusion analysis algorithm on the basis of neural network technology is proposed. This study designed a window adaptive strategy, use the probability density function, and built an impulse noise model, and use this model to divide each pixel in the image into noise points or signal points to obtain image impulse noise detection results, and use bilateral filtering algorithm to achieve image denoising processing; The fruit fly optimization algorithm is adopted to detect the edge of the multi sensor moving image, extract the image’s main edge points, and remove the detail edge points and noise points; nonlinear convolutional layer is used to replace most fusion layers to improve the dense network model, and the cross-entropy loss is used as the loss function in training the network, then use guided filters to generate guide maps, and generate final fusion images. According to experimental results, the noise detection method in this paper can also maintain 79.21% non-noise extraction rate under the noise density of 0.7. The highest correlation coefficient between the proposed algorithm and the standard image is 37.41. Its peak signal-to-noise ratio is as low as 0.09 and as high as 0.52. It has a minimum root mean square error of 8.52. The above values are better than other measured methods, and its edge miss rate can be as low as 1%, the image resolution is higher. It can be seen that its image denoising effect is better. Image denoising effect, and low edge missed detection rate, which effectively improves the effect of image fusion.

OutSingle: a novel method of detecting and injecting outliers in RNA-Seq count data using the optimal hard threshold for singular values.

Finding outliers in RNA-sequencing (RNA-Seq) gene expression (GE) can help in identifying genes that are aberrant and cause Mendelian disorders. Recently developed models for this task rely on modeling RNA-Seq GE data using the negative binomial distribution (NBD). However, some of those models either rely on procedures for inferring NBD's parameters in a nonbiased way that are computationally demanding and thus make confounder control challenging, while others rely on less computationally demanding but biased procedures and convoluted confounder control approaches that hinder interpretability. In this article, we present OutSingle (Outlier detection using Singular Value Decomposition), an almost instantaneous way of detecting outliers in RNA-Seq GE data. It uses a simple log-normal approach for count modeling. For confounder control, it uses the recently discovered optimal hard threshold (OHT) method for noise detection, which itself is based on singular value decomposition (SVD). Due to its SVD/OHT utilization, OutSingle's model is straightforward to understand and interpret. We then show that our novel method, when used on RNA-Seq GE data with real biological outliers masked by confounders, outcompetes the previous state-of-the-art model based on an ad hoc denoising autoencoder. Additionally, OutSingle can be used to inject artificial outliers masked by confounders, which is difficult to achieve with previous approaches. We describe a way of using OutSingle for outlier injection and proceed to show how OutSingle outperforms its competition on 16 out of 18 datasets that were generated from three real datasets using OutSingle's injection procedure with different outlier types and magnitudes. Our methods are applicable to other types of similar problems involving finding outliers in matrices under the presence of confounders. The code for OutSingle is available at https://github.com/esalkovic/outsingle.

Improving the forecasting accuracy of interval-valued carbon price from a novel multi-scale framework with outliers detection: An improved interval-valued time series analysis mode

Accurate carbon price forecasting can provide policymakers with a reasonable basis for carbon pricing. Interval-valued carbon price forecasting could provide sufficient information compared with real-valued carbon price time series prediction. On the other hand, current interval-valued carbon price forecasting has major challenges including data complexity, outliers, and the selection of forecasting methods, which make the forecasting results with great uncertainty and instability. To address these issues, this paper proposes an interval-valued carbon price forecasting method based on new data processing techniques, and discusses the effects of different combinations of interval variables on the forecasting results. We first established interval complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and interval outlier detection method (I-ksigma) to reduce the data complexity and identify outliers. Then, various neural network models include interval multilayer perceptron (IMLP), multi-output support vector regression (MSVR), long short-term memory network (LSTM), gated recurrent unit neural network (GRU), and convolution neural network (CNN) are chosen to conduct combination forecasting on the interval sub-modes produced by ICEEMDAN. The final results are obtained by summing the interval sub-modes. Finally, taking the carbon trading price in Hubei as the research object, the results show that the developed forecasting framework is superior to all comparison models in forecasting precision and stability. Furthermore, different combinations of interval variables (CRM, Minmax, L + 2R, and U-2R) lead to different decomposition results and outlier detection results, which finally affect the prediction results.