Information Gained Research Articles

Improving Fault Classification Accuracy Using Wavelet Transform and Random Forest with STATCOM Integration

INTRODUCTION: Fault detection in transmission lines is critical for keeping the grid stable and reliable. This research offers a new methodology, the Wavelet Transform-Enhanced Random Forest Fault Classification System with STATCOM Integration (WERFCS-SI), to solve the shortcomings of existing fault detection approaches. OBJECTIVES: The integration of STATCOM-compensated transmission lines improves fault detection capabilities. The Wavelet Transform finds faults by analysing approximation and detail coefficients, allowing for multiresolution analysis and exact fault localisation. METHODS: Feature selection approaches, such as information gain, are used to discover and keep relevant features, increasing classification accuracy. RESULTS: Due to its ability to process complex, high-dimensional data and identify minute feature connections, Random Forest (RF) is utilised for classification tasks. The proposed approach improves RF model performance while maintaining precision. CONCLUSION: The integrated technique simplifies fault categorisation, increasing accuracy and efficiency by detecting problems in the transmission line system.

Asteroid material classification based on multi-parameter constraints using artificial intelligence

Material types of asteroids provide key clues to their evolutionary history and contained resources. The Gaia mission has released extensive low-resolution spectral observation data of small Solar System bodies. However, methods for classifying asteroids based on low-resolution space-based spectra are still inadequate, and do not fully leverage the complementary features of spectra and multiple intrinsic attributes of asteroids to achieve precise material classification. Our goal is to propose a method with a higher generalization accuracy for asteroid material classification by integrating multi-source information, identifying optimal feature combinations for model inputs, and deepening the understanding of relationships among asteroid parameters. The effective asteroid photometric, physical, and orbital parameters were screened using the information gain ratio and Spearman's rank correlation coefficient. Then, artificial intelligence techniques were employed to combine asteroid spectra with the selected various parameters for six-class material classification. By comparing five machine learning models, we identified network structures with higher validation accuracy and stable generalization performance. Meanwhile, feature ablation experiments were conducted to determine the input parameter combinations suitable for different scenarios. Finally, based on the statistical results and model outputs, the constraint relationships among asteroid parameters were visualized and analyzed. The proposed AsterRF model achieved a validation accuracy of 92.2<!PCT!>, an improvement of approximately 7.8 percentage points compared to existing methods that use only spectra. V-type asteroids exhibited the highest classification accuracy, followed by A-type and D-type. X-type asteroids had the lowest precision and recall, and were easily confused with C-type. The model generally showed higher classification confidence for S-type asteroids. The top five attributes that the model focused on are the phase slope parameter (G), orbital type, albedo, H magnitude, and effective diameter. Additionally, the correlations between asteroid materials and other parameters were generally below 0.4. Incorporating optimal asteroid parameter combinations can significantly enhance classification accuracy based on spectra. A dual-channel network that processes spectra and parameter inputs separately, and employs a self-attention mechanism for feature fusion is effective in combining multi-source asteroid information. Both the statistical correlations and model performance-based importance rankings of parameters contribute to understanding the constraint relationships among asteroid attributes.

EMSIG: Uncovering Factors Influencing COVID-19 Vaccination Across Different Subgroups Characterized by Embedding-Based Spatial Information Gain

Background/Objectives: COVID-19 and its variants continue to pose significant threats to public health, with considerable uncertainty surrounding their impact. As of September 2024, the total number of deaths reached 8.8 million worldwide. Vaccination remains the most effective strategy for preventing COVID-19. However, vaccination rates in the Deep South, U.S., are notably lower than the national average due to various factors. Methods: To address this challenge, we developed the Embedding-based Spatial Information Gain (EMSIG) method, an innovative tool using machine learning techniques for subgroup modeling. EMSIG helps identify subgroups where participants share similar perceptions but exhibit high variance in COVID-19 vaccine doses. It introduces spatial information gain (SIG) to screen regions of interest (ROI) subgroups and reveals their specific concerns. Results: We analyzed survey data from 1020 participants in Alabama. EMSIG identified 16 factors encompassing COVID-19 hesitancy and trust in medical doctors, pharmacists, and public health authorities and revealed four distinct ROI subgroups. The five factors, including COVID-19 perceived detriment, fear, skepticism, side effects related to COVID-19, and communication with pharmacists, were commonly shared across at least three subgroups. A subgroup primarily composed of Democrats with a high flu-shot rate expressed concerns about pharmacist communication, government fairness, and responsibility. Another subgroup, characterized by older, white Republicans with a relatively low flu-shot rate, expressed concerns about doctor trust and the intelligence of public health authorities. Conclusions: EMSIG enhances our understanding of specific concerns across different demographics, characterizes these demographics, and informs targeted interventions to increase vaccination uptake and ensure equitable prevention strategies.

Optimal experimental design for identification of hydrodynamic loading models

Abstract Hydrodynamic model-scale experiments are an intrinsic part of the design of marine structures, as they enable validating and calibrating the involved hydrodynamic numerical models. Such seakeeping experiments are generally conducted using a simple spring-based mooring system, with fixed properties throughout the tests. In the context of cyber-physical testing, the marine structure is kept in position in the laboratory by a virtual mooring system, the properties of which can be adjusted on the fly. This paper provides an algorithmic approach for maximizing the information gain from the test by optimizing the mooring stifness and damping parameters that define the experiment. The method is verified with synthetic data generated for the INO Windmoor 12MW floater, and is shown to be robust to noise and unmodeled effects.

Research and performance analysis of random forest-based feature selection algorithm in sports effectiveness evaluation

The rapid progress in fields such as data mining and machine learning, as well as the explosive growth of sports big data, have posed new challenges to the research of sports big data. Most of the available sports data mining techniques concentrates on extracting and constructing effective features for basic sports data, which cannot be achieved simply by using data statistics. Especially in the targeted mining of sports data, traditional mining techniques still have shortcomings such as low classification accuracy and insufficient refinement. In order to solve the problem of low accuracy in traditional mining methods, the study combines the random forest algorithm with the artificial raindrop algorithm, and adopts a sports data mining method based on feature selection to achieve effective analysis of sports big data. This study is based on the evaluation method of motion effects using random forests, and uses feature extraction algorithms to study the motion effect impacts. It uses the information gain index to rank the importance of features and accurately gain the degree of influence of exercise on various indicators of the human body. Through simulation verification, the algorithm proposed by the research institute performs the best in accuracy and FI scores on the training and testing sets, with accuracies of 0.849 ± 0.021 and 0.819 ± 0.022, respectively, and F1 scores of 0.837 ± 0.020 and 0.864 ± 0.021, respectively. This indicates that the algorithm proposed by the research institute has high classification accuracy and performance proves that the Random Forest-based feature selection algorithm established in this study is superior to the existing traditional feature extraction and extraction methods in terms of both performance and accuracy. The proposal of this data analysis method has achieved accurate and efficient utilization of sports big data, which is of great significance for the development of the sports education industry.

Experimental of information gain and AdaBoost feature for machine learning classifier in media social data

Experimental of information gain and AdaBoost feature for machine learning classifier in media social data

Less is More: Unlocking Semi-Supervised Deep Learning for Vulnerability Detection

Deep learning has demonstrated its effectiveness in software vulnerability detection, but acquiring a large number of labeled code snippets for training deep learning models is challenging due to labor-intensive annotation. With limited labeled data, complex deep learning models often suffer from overfitting and poor performance. To address this limitation, semi-supervised deep learning offers a promising approach by annotating unlabeled code snippets with pseudo-labels and utilizing limited labeled data together as training sets to train vulnerability detection models. However, applying semi-supervised deep learning for accurate vulnerability detection comes with several challenges. One challenge lies in how to select correctly pseudo-labeled code snippets as training data, while another involves mitigating the impact of potentially incorrectly pseudo-labeled training code snippets during model training. To address these challenges, we propose the Semi-Supervised Vulnerability Detection (SSVD) approach. SSVD leverages the information gain of model parameters as the certainty of the correctness of pseudo-labels and prioritizes high-certainty pseudo-labeled code snippets as training data. Additionally, it incorporates the proposed noise-robust triplet loss to maximize the separation between vulnerable and non-vulnerable code snippets to better propagate labels from labeled code snippets to nearby unlabeled snippets, and utilizes the proposed noise-robust cross-entropy loss for gradient clipping to mitigate the error accumulation caused by incorrect pseudo-labels. We evaluate SSVD with nine semi-supervised approaches on four widely-used public vulnerability datasets. The results demonstrate that SSVD outperforms the baselines with an average of 29.82% improvement in terms of F1-score and 56.72% in terms of MCC. In addition, SSVD trained on a certain proportion of labeled data can outperform or closely match the performance of fully supervised LineVul and ReVeal vulnerability detection models trained on 100% labeled data in most scenarios. This indicates that SSVD can effectively learn from limited labeled data to enhance vulnerability detection performance, thereby reducing the effort required for labeling a large number of code snippets.

Machine learning for air quality index (AQI) forecasting: shallow learning or deep learning?

In this study, several machine learning (ML) models consisting of shallow learning (SL) models (e.g., random forest (RF), K-nearest neighbor (KNN), weighted K-nearest neighbor (WKNN), support vector machine (SVM), artificial neural network (ANN), and deep learning (DL) models (e.g., long short-term memory (LSTM), gated recurrent unit (GRU), recurrent neural network (RNN), and convolutional neural network (CNN)) have been employed for predicting air pollution and its classification. The models were selected based on factors such as prediction accuracy, model generalization, model complexity, and training time. Our study focuses on analyzing and predicting the air quality index (AQI) using daily PM10 concentration as natural pollutants and nine meteorological parameters from March 2013 to February 2022 in Zabol. We also utilized the information gain (IG) method for feature selection. Several measures including accuracy, F1 score, precision, recall, and the area under the curve (AUC), are computed to assess model performance. This study demonstrates the efficacy of DL models, particularly CNN, in predicting the AQI with remarkable accuracy. Our findings reveal that all models effectively classify air quality levels, with an AUC of 0.95 for the good class in both DL and ANN models, significantly outperforming SL models. The AUC values for the hazardous and moderate classes of DL models were also impressive, at 0.90 and 0.83, respectively, underscoring their effectiveness in critical classifications. In terms of performance, CNN achieved an accuracy of 0.60, leading the models, while RF followed closely at 0.58. RNN, GRU, ANN, and SVM each reached an accuracy of 0.57, demonstrating a competitive edge. LSTM and WKNN recorded an accuracy of 0.55, and KNN was slightly lower at 0.53. These results highlight the superior capabilities of DL models in addressing complex air quality classifications, providing invaluable insights for policymakers. By leveraging these advanced techniques, stakeholders can implement more effective strategies to combat air pollution and safeguard public health. It is worth noting that irregular monitoring of air quality data may affect the robustness of our predictions, highlighting the need for more consistent data collection to ensure an accurate representation of pollution levels.

Assessment of land degradation susceptibility within the Shaqlawa subregion of Northern Iraq-Kurdistan Region via synergistic application of remotely acquired datasets and advanced predictive models.

Land degradation (LD) is the decline in a land's functional capacity and productive potential, which includes various anthropogenic and natural drivers. This study focuses on three primary manifestations of LD including soil erosion, landslides, and rockfalls, which are the most prevalent in the Shaqlawa district. A set of 22 LD conditioning factors, encompassing curvature, lithology, aspect, river density, soil type, lineament density, river distance, elevation, road distance, length slope (LS), land use land cover (LULC), stream power index (SPI), valley depth, profile curvature, slope, solar radiation, road density, lineament distance, rainfall, topographic wetness index (TWI), plan curvature, and normalized difference vegetation index (NDVI), were integrated into the analysis. Variance inflation factors (VIF) and tolerance (TOL) values from linear regression indicate that most LD factors have acceptable levels of multicollinearity. The Information Gain Ratio (IGR) identified key variables TWI, NDVI, and lithology-as pivotal factors for predicting LD. Additionally, the study evaluated degradation factors using various machine learning (ML) algorithms, including random forest (RF), Naive Bayes, logistic regression, rotation forest, forest penalized attributes (FPA), and Fisher's Linear discriminant analysis (FLDA). This facilitated categorizing the study area into five susceptibility categories. The FLDA model categorized the highest area under very high degradation risk at 26.72%, emphasizing the varied insights each algorithm brought to characterizing the degradation risk. Additionally, the receiver operating characteristic curves (ROC) were employed for model validation, identifying RF as the most successful model in the training dataset with an area under the curve (AUC) of 0.882, while FLDA outperformed in the testing dataset with an AUC of 0.883. The identified LD-prone areas will help land-use planners and emergency management officials apply effective mitigation strategies for similar terrains.

An Improved Design for a Cloud Intrusion Detection System Using Hybrid Features Selection Approach with ML Classifier

The rapid adoption of cloud computing has introduced new security concerns, particularly regarding data privacy and protection from sophisticated cyberattacks. Intrusion Detection Systems (IDSs) based on Machine Learning (ML) offer improved detection of malicious behaviour in network traffic. However, existing IDS solutions struggle with large-dimensional data and imbalanced datasets, resulting in higher false positive rates (FPR) and reduced detection rates (DR). In this paper, we propose an improved Cloud IDS leveraging a hybrid features selection approach using Information Gain (IG), Chi-square (CS), and Particle Swarm Optimization (PSO). To address the challenge of imbalanced datasets, the Synthetic Minority Over-sampling Technique (SMOTE) is employed. The Random Forest (RF) classifier is used to detect and classify attacks. The system is evaluated on the UNSW-NB15 and Kyoto datasets, achieving accuracies of over 98% and 99%, respectively, outperforming other methods in terms of detection accuracy.

LIMIT: Learning Interfaces to Maximize Information Transfer

Robots can use auditory, visual, or haptic interfaces to convey information to human users. The way these interfaces select signals is typically pre-defined by the designer: for instance, a haptic wristband might vibrate when the robot is moving and squeeze when the robot stops. But different people interpret the same signals in different ways, so that what makes sense to one person might be confusing or unintuitive to another. In this article, we introduce a unified algorithmic formalism for learning co-adaptive interfaces from scratch . Our method does not need to know the human’s task (i.e., what the human is using these signals for). Instead, our insight is that interpretable interfaces should select signals that maximize correlation between the human’s actions and the information the interface is trying to convey. Applying this insight we develop Learning Interfaces to Maximize Information Transfer (LIMIT). LIMIT optimizes a tractable, real-time proxy of information gain in continuous spaces. The first time a person works with our system the signals may appear random; but over repeated interactions, the interface learns a one-to-one mapping between displayed signals and human responses. Our resulting approach is both personalized to the current user and not tied to any specific interface modality. We compare LIMIT to state-of-the-art baselines across controlled simulations, an online survey, and an in-person user study with auditory, visual, and haptic interfaces. Overall, our results suggest that LIMIT learns interfaces that enable users to complete the task more quickly and efficiently, and users subjectively prefer LIMIT to the alternatives. See videos here: https://youtu.be/IvQ3TM1_2fA .

Clarity in complexity: how aggregating explanations resolves the disagreement problem

The Rashômon Effect, applied in Explainable Machine Learning, refers to the disagreement between the explanations provided by various attribution explainers and to the dissimilarity across multiple explanations generated by a particular explainer for a single instance from the dataset (differences between feature importances and their associated signs and ranks), an undesirable outcome especially in sensitive domains such as healthcare or finance. We propose a method inspired from textual-case based reasoning for aligning explanations from various explainers in order to resolve the disagreement and dissimilarity problems. We iteratively generated a number of 100 explanations for each instance from six popular datasets, using three prevalent feature attribution explainers: LIME, Anchors and SHAP (with the variations Tree SHAP and Kernel SHAP) and consequently applied a global cluster-based aggregation strategy that quantifies alignment and reveals similarities and associations between explanations. We evaluated our method by weighting the \\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\:k$$\\end{document}-NN algorithm with agreed feature overlap explanation weights and compared it to a non-weighted \\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\:k$$\\end{document}-NN predictor, having as task binary classification. Also, we compared the results of the weighted \\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\:k$$\\end{document}-NN algorithm using aggregated feature overlap explanation weights to the weighted \\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\:k$$\\end{document}-NN algorithm using weights produced by a single explanation method (either LIME, SHAP or Anchors). Our global alignment method benefited the most from a hybridization with feature importance scores (information gain), that was essential for acquiring a more accurate estimate of disagreement, for enabling explainers to reach a consensus across multiple explanations and for supporting effective model learning through improved classification performance.

Assessing critical flood-prone districts and optimal shelter zones in the Brahmaputra Valley: Strategies for effective flood risk management

Frequent flooding has become a persistent issue in floodplain regions, causing significant disasters during each rainy season due to insufficient disaster management planning. This study proposes a methodology to prioritize flood susceptibility areas at the district level and identify suitable sites for flood shelters using a combination of machine learning algorithms and multi-criteria analysis, supported by geospatial technology. Flood shelter suitability mapping was conducted using the Analytical Hierarchy Process (AHP), while flood susceptibility zones were assessed using four different machine learning models: Support Vector Machine (SVM), Random Forest, Decision Tree, and Naive Bayes. The integration of machine learning models with the AHP technique is vital in situations where conventional numerical models face challenges due to limited data, such as river discharge and water levels. The methodology includes a multicollinearity assessment to ensure the independence of selected flood-causing factors, information gain ratio to identify the most influential factors, Spearman's rho test to verify correlations between the machine learning models, and ROC-AUC along with statistical regression for validating the accuracy of the flood susceptibility maps. The findings indicate that the SVM algorithm, given its strong performance and effective training datasets, is recommended for areas with similar physical characteristics. The district-wise priority map generated from the weighted results of flood susceptibility assessments will be useful for flood management and mitigation strategies. Additionally, the study found that applying the AHP technique to determine flood shelter suitability, after assessing flood-prone areas, enhanced the efficiency of the flood management process. This research offers valuable insights for authorities to better address flooding and improve flood prevention and management efforts in floodplain regions, contributing to broader climate change adaptation strategies.

A rate-them-all lineup procedure increases information but reduces discriminability.

Prior research has investigated ways to optimize identification performance, but an open question concerns exactly what variable should be optimized. One reasonable way to optimize performance is to maximize discriminability, which is achieved by increasing correct identifications of guilty suspects while simultaneously decreasing false identifications of innocent suspects. Another reasonable way to optimize performance is to maximize the information about the guilt or innocence of the suspect, which is best achieved by ensuring that a confidence rating is always made to the suspect. In a typical lineup, however, limited information about the suspect is obtained if the witness picks a filler or rejects the lineup. One proposed solution to that problem is to have the witness provide a confidence rating to every member of the lineup (a rate-them-all lineup). But what effect, if any, does a rate-them-all procedure have on discriminability? To answer that question, we compared a rate-them-all lineup procedure to standard simultaneous lineup and showup procedures using receiver operating characteristic analysis. In terms of discriminability, the rate-them-all procedure was diagnostically inferior to both. A reasonable goal for future research is to make use of theoretical models of eyewitness identification to simultaneously maximize both discriminability and information gain. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Efficient estimation of expected information gain in Bayesian experimental design with multi-index Monte Carlo

Efficient estimation of expected information gain in Bayesian experimental design with multi-index Monte Carlo

ShNFN: Shepard Neuro-Fuzzy Network for Intrusion Detection in Fog Computing

Fog computing is a type of distributed computing that makes data storage and computation closer to the network edge. While fog computing offers numerous advantages, it also introduces several challenges, particularly in terms of security. Intrusion Detection System (IDS) plays a crucial role in securing fog computing environments by monitoring network traffic and system activities for signs of malicious behavior. Several techniques can be employed to enhance intrusion detection in fog computing environments. Accordingly, this paper proposes a Shepard Neuro-Fuzzy Network (ShNFN) for intrusion detection in fog computing. Initially, in the cloud layer, the input data are passed to data transformation to transform the unstructured data into structured form. Here, data transformation is done employing the Box-Cox transformation. Following this, the feature selection is done in terms of information gain and symmetric uncertainty process and it is used to create a relationship between two variables. After that, the data are classified by employing the proposed ShNFN. The ShNFN is attained by fusing two networks, such as Cascade Neuro-Fuzzy Network (Cascade NFN) and Shepard Convolutional Neural Networks (ShCNN). After this, the physical process is executed at the endpoint layer. Finally, intrusion detection is accomplished in the fog layer by the proposed ShNFN method. The performance of the intrusion detection using ShNFN is calculated by the metrics of recall, F-measure and precision. The proposed method achieves the values of 93.3%, 92.5% and 94.8% for recall, F-measure, and precision, respectively.

That's the way I (dis-)like it! - Effect of Gamer Types and Competitiveness in VR Exergaming

In response to growing global concerns surrounding sedentary lifestyles among adults, the imperative to motivate physical activity has driven the exploration of Virtual Reality (VR) exergames as a promising solution. Personality traits, particularly competitiveness, can play a pivotal role in influencing individuals' motivation to engage in either competitive or non-competitive versions of the same exergame. While existing studies have primarily focused on sports-related competitive orientations, they have overlooked the potential gain in information from gaming-related personality traits. This research addresses this gap by using a modified gaming-related competitiveness questionnaire and evaluating a VR cycling exergame featuring solo, self-competition and virtual opponent cycling conditions in a controlled user study with 38 participants. Empirical findings reveal a notable difference in self-rated competitiveness between the two domains of sports and gaming, and significant correlations between some gamer-types and motivation. These findings highlight the importance of these personality types during exergaming and their potential efficacy for personalisation to increase motivation.

Predicting Dependent Edges in Nonequilibrium Complex Systems Based on Overlapping Module Characteristics

Problem: Predicting dependency relationships in nonequilibrium systems is a critical challenge in complex systems research. Solution proposed: In this paper, we propose a novel method for predicting dependent edges in network models of nonequilibrium complex systems, based on overlapping module features. This approach addresses the many-to-many dependency prediction problem between nonequilibrium complex networks. By transforming node-based network models into edge-based models, we identify overlapping modular structures, enabling the prediction of many-to-many dependent edges. Experimental evaluation: This method is applied to dependency edge prediction in power and gas networks, curriculum and competency networks, and text and question networks. Results: The results indicate that the proposed dependency edge prediction method enhances the robustness of the network in power–gas networks, accurately identifies supporting relationships in curriculum–competency networks, and achieves better information gain in text–question networks. Conclusion: These findings confirm that the overlapping module-based approach effectively predicts dependencies across various nonequilibrium complex systems in diverse fields.

Should the multi-layer transportation network structure be reduced?

ABSTRACT The increasing diversity of transportation modes and the rapid expansion of transportation networks present significant challenges for modeling multi-layer comprehensive transportation networks. It is crucial to determine whether aggregating certain layers is a viable option for balancing complexity reduction and information preservation. This decision defines the layered structures and informs subsequent analyses of these networks. Two-dimensional factors, namely topological structures and transportation attributes, are considered to enhance understanding of the similarities among network layers. The relative entropy and the Gini index are employed as metrics to assess information gain or loss resulting from layer aggregation or segregation, guiding decisions on network reduction. Furthermore, an integrated similarity measure, based on the quantum Jensen-Shannon divergence and the Gower distance, is utilized to identify the optimal aggregation sequences. Two real-world transportation networks serve as case studies. Results demonstrate that these transportation networks are more effectively maintained with layer-separated structures, preserving maximum information.

ECG signal classification based on DWT denoising and XGBoost

Electrocardiogram signal (ECG) can directly reflect the health status of the heart, and is an important basis for prevention and treatment of heart disease. In order to realize ECG signal classification effectively, an ECG signal classification method based on discrete wavelet transform and Xgboost is proposed in this paper, which improves the accuracy of ECG signal classification. Specifically, we first divide, select and downsample the heart beat of the data, and then use the discrete wavelet transform to reduce the noise of the data set to improve the signal to noise ratio. Finally, we use Xgboost algorithm as the classifier to classify the data, and get 98.7% accuracy rate on the test set. In each module, we carried out comparative experiments to verify the correctness and rigor of our method. In addition, in order to make up for the lack of interpretability of traditional machine learning methods, we defined the importance of each feature according to the information gain generated by different features to the model during the training of XGBoost, and then got the key bands that should be paid attention to when distinguishing heart beats, which improved the interpretability of the model. It also provides a scientific basis for the classification of ECG signals and practical medical work.