Single Classification Algorithm Research Articles

A novel hybrid supervised and unsupervised hierarchical ensemble for COVID-19 cases and mortality prediction

Though COVID-19 is no longer a pandemic but rather an endemic, the epidemiological situation related to the SARS-CoV-2 virus is developing at an alarming rate, impacting every corner of the world. The rapid escalation of the coronavirus has led to the scientific community engagement, continually seeking solutions to ensure the comfort and safety of society. Understanding the joint impact of medical and non-medical interventions on COVID-19 spread is essential for making public health decisions that control the pandemic. This paper introduces two novel hybrid machine-learning ensembles that combine supervised and unsupervised learning for COVID-19 data classification and regression. The study utilizes publicly available COVID-19 outbreak and potential predictive features in the USA dataset, which provides information related to the outbreak of COVID-19 disease in the US, including data from each of 3142 US counties from the beginning of the epidemic (January 2020) until June 2021. The developed hybrid hierarchical classifiers outperform single classification algorithms. The best-achieved performance metrics for the classification task were Accuracy = 0.912, ROC-AUC = 0.916, and F1-score = 0.916. The proposed hybrid hierarchical ensemble combining both supervised and unsupervised learning allows us to increase the accuracy of the regression task by 11% in terms of MSE, 29% in terms of the area under the ROC, and 43% in terms of the MPP metric. Thus, using the proposed approach, it is possible to predict the number of COVID-19 cases and deaths based on demographic, geographic, climatic, traffic, public health, social-distancing-policy adherence, and political characteristics with sufficiently high accuracy. The study reveals that virus pressure is the most important feature in COVID-19 spread for classification and regression analysis. Five other significant features were identified to have the most influence on COVID-19 spread. The combined ensembling approach introduced in this study can help policymakers design prevention and control measures to avoid or minimize public health threats in the future.

Comparative study of different machine learning models in landslide susceptibility assessment: A case study of Conghua District, Guangzhou, China

Machine learning is currently one of the research hotspots in the field of landslide prediction. To clarify and evaluate the differences in characteristics and prediction effects of different machine learning models, Conghua District, which is the most prone to landslide disasters in Guangzhou, was selected for landslide susceptibility evaluation. The evaluation factors were selected by using correlation analysis and variance expansion factor method. Applying four machine learning methods namely Logistic Regression (LR), Random Forest (RF), Support Vector Machines (SVM), and Extreme Gradient Boosting (XGB), landslide models were constructed. Comparative analysis and evaluation of the model were conducted through statistical indices and receiver operating characteristic (ROC) curves. The results showed that LR, RF, SVM, and XGB models have good predictive performance for landslide susceptibility, with the area under curve (AUC) values of 0.752, 0.965, 0.996, and 0.998, respectively. XGB model had the highest predictive ability, followed by RF model, SVM model, and LR model. The frequency ratio (FR) accuracy of LR, RF, SVM, and XGB models was 0.775, 0.842, 0.759, and 0.822, respectively. RF and XGB models were superior to LR and SVM models, indicating that the integrated algorithm has better predictive ability than a single classification algorithm in regional landslide classification problems.©2024 China Geology Editorial Office.

An ensemble model to detect packet length covert channels

<span lang="EN-US">Covert channel techniques have enriched the way to commit dangerous and unwatched attacks. They exploit ways that are not intended to convey information; therefore, traditional security measures cannot detect them. One class of covert channels that difficult to detect, mitigate, or eliminate is packet length covert channels. This class of covert channels takes advantage of packet length variations to convey covert information. Numerous research articles reflect the useful use of machine learning (ML) classification approaches to discover covert channels. Therefore, this study presented an efficient ensemble classification model to detect such types of attacks. The ensemble model consists of five machine learning algorithms representing the base classifiers. The base classifiers include naive Bayes (NB), decision tree (DT), support vector machine (SVM), k-nearest neighbor (KNN), and random forest (RF). Whereas, the logistic regression (LR) classifier was employed to aggregate the outputs of the base classifiers and thus to generate the ensemble classifier output. The results showed a good performance of our proposed ensemble classifier. It beats all single classification algorithms, with a 99.3% accuracy rate and negligible classification errors.</span>

GA-StackingMD: Android Malware Detection Method Based on Genetic Algorithm Optimized Stacking

With the rapid development of network and mobile communication, intelligent terminals such as smartphones and tablet computers have changed people’s daily life and work. However, malware such as viruses, Trojans, and extortion applications have introduced threats to personal privacy and social security. Malware of the Android operating system has a great variety and updates rapidly. Android malware detection is faced with the problems of high feature dimension and unsatisfied detection accuracy of single classification algorithms. In this work, an Android malware detection framework GA-StackingMD is presented, which employs Stacking to compose five different base classifiers, and Genetic Algorithm is applied to optimize the hyperparameters of the framework. Experiments show that Stacking could effectively improve malware detection accuracy compared with single classifiers. The presented GA-StackingMD achieves 98.43% and 98.66% accuracies on CIC-AndMal2017 and CICMalDroid2020 data sets, which shows the effectiveness and feasibility of the proposed method.

Review on IDS based on ML Algorithms

Abstract: Intrusion detection is one of the challenging problems encountered by the modern network security industry. The developing pace of digital assaults on framework networks as of late compounds the protection and security of PC foundation and PCs. Intrusion Detection and Prevention systems are transforming into a critical part of PC organizations and network safety. Various approaches have been proposed to determine the most effective features and hence enhance the efficiency of intrusion detection systems, the methods include, machine learning-based (ML), Bayesian based algorithm, Random Forest, SVM, Decision Tree. This paper presents an intensive survey on different examination articles that utilized single, hybrid and ensemble classification algorithms. The outcomes measurements, weaknesses and datasets involved by the concentrated on articles in the advancement of IDS were looked at. A future heading for potential explores is likewise given. The paper addressed latest research papers written from the use of machine learning classifiers in intrusion detection systems.

Detecting diseases in medical prescriptions using data mining methods

Every year, the health of millions of people around the world is compromised by misdiagnosis, which sometimes could even lead to death. In addition, it entails huge financial costs for patients, insurance companies, and governments. Furthermore, many physicians’ professional life is adversely affected by unintended errors in prescribing medication or misdiagnosing a disease. Our aim in this paper is to use data mining methods to find knowledge in a dataset of medical prescriptions that can be effective in improving the diagnostic process. In this study, using 4 single classification algorithms including decision tree, random forest, simple Bayes, and K-nearest neighbors, the disease and its category were predicted. Then, in order to improve the performance of these algorithms, we used an Ensemble Learning methodology to present our proposed model. In the final step, a number of experiments were performed to compare the performance of different data mining techniques. The final model proposed in this study has an accuracy and kappa score of 62.86% and 0.620 for disease prediction and 74.39% and 0.720 for prediction of the disease category, respectively, which has better performance than other studies in this field.In general, the results of this study can be used to help maintain the health of patients, and prevent the wastage of the financial resources of patients, insurance companies, and governments. In addition, it can aid physicians and help their careers by providing timely information on diagnostic errors. Finally, these results can be used as a basis for future research in this field.

An Efficient Method Combined Data-Driven for Detecting Electricity Theft with Stacking Structure Based on Grey Relation Analysis

Nowadays, electricity theft has been a major problem worldwide. Although many single-classification algorithms or an ensemble of single learners (i.e., homogeneous ensemble learning) have proven able to automatically identify suspicious customers in recent years, after the accuracy of these methods reaches a certain level, it still cannot be improved even if it continues to be optimized. To break through this bottleneck, a heterogeneous ensemble learning method with stacking integrated structure of different strong individual learners for detection of electricity theft is presented in this paper. Firstly, we use the grey relation analysis (GRA) method to select the heterogeneous strong classifier combination of LG + LSTM + KNN as the base model layer of stacking structure based on the principle of the highest comprehensive evaluation index value. Secondly, the support vector machine (SVM) model with relatively good results of the stacking overall structure experiment is selected as the model of the meta-model layer. In this way, a heterogeneous integrated learning model for electricity theft detection of the stacking structure is constructed. Finally, the experiments of this model are conducted on electricity consumption data from State Grid Corporation of China, and the results show that the detection performance of the proposed method is better than that of the existing state-of-the-art detection method (where the area under receiver operating characteristic curve (AUC) value is 0.98675).

An approach for improved students’ performance prediction using homogeneous and heterogeneous ensemble methods

<span lang="EN-US">Web-based learning technologies of educational institutions store a massive amount of interaction data which can be helpful to predict students’ performance through the aid of machine learning algorithms. With this, various researchers focused on studying ensemble learning methods as it is known to improve the predictive accuracy of traditional classification algorithms. This study proposed an approach for enhancing the performance prediction of different single classification algorithms by using them as base classifiers of homogeneous ensembles (bagging and boosting) and heterogeneous ensembles (voting and stacking). The model utilized various single classifiers such as multilayer perceptron or neural networks (NN), random forest (RF), naïve Bayes (NB), J48, JRip, OneR, logistic regression (LR), k-nearest neighbor (KNN), and support vector machine (SVM) to determine the base classifiers of the ensembles. In addition, the study made use of the University of California Irvine (UCI) open-access student dataset to predict students’ performance. The comparative analysis of the model’s accuracy showed that the best-performing single classifier’s accuracy increased further from 93.10% to 93.68% when used as a base classifier of a voting ensemble method. Moreover, results in this study showed that voting heterogeneous ensemble performed slightly better than bagging and boosting homogeneous ensemble methods.</span>

A discrimination model for differentiation of renal cell carcinoma from renal angiomyolipoma without visible fat: based on hierarchical fusion framework of multi-classifier

To investigate the capabilities of classification models based on hierarchical fusion framework of multi-classifier using a random projection strategy for differentiation of renal cell carcinoma (RCC) from small renal angiomyolipoma (< 4 cm) without visible fat (AMLwvf). We retrospectively collected the clinical data from 163 patients with pathologically proven small renal mass, including 118 with RCC and 45 with AMLwvf.Target region of interest (ROI) delineation was performed on an unenhanced phase (UP) CT image slice displaying the largest lesion area.The radiomics features were used to establish a hierarchical fusion method.On the projection-based level, the homogeneous classifiers were fused, and the fusion results were further fused at the classifier-based level to construct a multi-classifier fusion system based on random projection for differentiation of AMLwvf and RCC.The discriminative capability of this model was quantitatively evaluated using 5-fold cross validation and 4 evaluation indexes[specificity, sensitivity, accuracy and area under ROC curve (AUC)].We quantitatively compared this multi-classifier fusion framework against different classification models using a single classifier and several multi-classifier ensemble models. When the projection number was set at 10, the proposed hierarchical fusion differentiation framework achieved the best results on all the evaluation measurements.At the optimal projection number of 10, the specificity, sensitivity, average accuracy and AUC of the multi-classifier ensemble classification system for differentiation between AMLwvf and RCC were 0.853, 0.693, 0.809 and 0.870, respectively. The proposed model constructed based on a multi-classifier fusion system using random projection shows better performance to differentiate RCC from AMLwvf than the AMLwvf and RCC discrimination models based on a single classification algorithm and the currently available benchmark ensemble methods.

Dual-Layer Deep Ensemble Techniques for Classifying Heart Disease

The prevalence of heart disease is increasing at a rapid rate due to changes in food habits and lifestyle of peopleall over the world. Early prediction and diagnosis of this fatal disease is a highly excruciating task. Nowadays, theensemble learning approaches are preferred owing to their effectiveness in performance when compared to theperformance of a single classification algorithm. In this work, a Dual-Layer Stacking Ensemble (DLSE) techniqueand a Deep Heterogeneous Ensemble (DHE) technique to classify heart disease are proposed. The DLSE uses several heterogeneous classifiers to form an ensemble that is efficient as well as diverse. The proposed frameworkconsists of two layers with the first layer consisting of three different base learning algorithms Naïve Bayes (NB),Decision Tree (DT), and Support Vector Machine (SVM). The second layer comprises of three different classifiers, Extremely Randomized Trees (ERT), Ada Boost Classifier (ABC) and Random Forest (RF). The second layerutilizes the results from the first layer to provide a diverse input for the three classifiers. Finally, the outcomesare fed to the meta-classifier Gradient Boosted Trees (GBT) to generate the final prediction. The DHE uses threedeep learning models Convolutional Neural Networks with Bidirectional Long Short-Term Memory (CNN BiLSTM), Artificial Neural Network (ANN) and Recurrent Neural Network (RNN) with RF, ERT and GBT as themeta-learners. The performance of the proposed methods is compared with traditional state-of-the-art classifiersas well as existing ensemble learning and deep learning methods. The experimental outcomes show that the proposed DLSE and DHE methods perform exceptionally in terms of accuracy, precision and recall measures.

Single kernel aflatoxin and fumonisin contamination distribution and spectral classification in commercial corn

Aflatoxin and fumonisin contamination distribution in corn is non-homogeneous. Therefore, bulk sample testing may not accurately represent the levels of contamination. Single kernel analysis could provide a solution to these problems and lead to remediation strategies such as sorting. Our study uses extensive single kernel aflatoxin (AF) and fumonisin (FM) measurements to (i) demonstrate skewness, calculate weighted sums of toxin contamination for a sample, and compare those values to bulk measurements, and (ii) improve single kernel classification algorithm performance. Corn kernels with natural contamination of aflatoxin and fumonisin (n = 864, from 9 bulk samples) were scanned individually twice for reflectance between the ultraviolet–visible–near infrared spectrum (304 nm–1086 nm), then ground and measured for aflatoxin and fumonisin using ELISA. Single kernel contamination distribution was non-homogeneous with 1.0% (n = 7) of kernels with ≥20 ppb aflatoxin (range 0 - 4.2×105 ppb), and 5.0% (n = 45) kernels with ≥2 ppm fumonisin (range 0 - 7.0×102 ppm). A single kernel weighted sum was calculated and compared to bulk measurements. Average difference in mycotoxin levels (AF = 0.0 log(ppb), FM = 0.0 log(ppm), weighted sum – measured bulk levels) calculated no systematic bias between the two methods, though with considerable range of −1.4 to 0.7 log(ppb) for AF and −0.6 to 0.8 log(ppm) for FM. Algorithms were trained on 70% of the kernels to classify aflatoxin (≥20ppb) and fumonisin (≥2ppm), while the remaining 30% of kernels were used for testing. For aflatoxin, the best performing algorithm was stochastic gradient boosting model with an accuracy of 0.83 (Sensitivity (Sn) = 0.75, Specificity (Sp) = 0.83), for both training and testing set. For fumonisin, the penalized discriminant analysis outperformed the rest of the algorithms, with a training accuracy of 0.89 (Sn = 0.87, Sp = 0.88), and testing accuracy of 0.86 (Sn = 0.78, Sp = 0.87). The present study improves the foundations for single kernel classification of aflatoxin and fumonisin in corn, and can be applied to high throughput screening. This study demonstrates the heterogeneous distribution of aflatoxin and fumonisin contamination at single kernel level, comparing bulk levels calculated from those data to traditional bulk tests, and utilizing a UV–Vis–NIR spectroscopy system to classify single corn kernels by aflatoxin and fumonisin level.

A comparative study and performance analysis of multirelational classification algorithms

Classification is one of the important tasks in data mining in which a model is generated-based on training dataset and that model is used to predict class label of unknown dataset. Many propositional classification algorithms exist to build accurate and scalable classifiers, applied to single table dataset only. Most real-world data are structured and stored in relational format and single table classification algorithms that cannot deal directly with relational data. Hence, the need for a multirelational classification algorithm that learns relational data and predicts class labels for relational tuple arises. For relational classification, various techniques are available that include flattening relational data, upgrading existing algorithm, and multiview learning. This paper presents comparative analysis of these techniques and algorithms in detail and shows that multiview-based algorithms outperform other algorithms. By implementing multiview-based algorithms it demonstrated that these algorithms achieve higher accuracy for binary class classification than multiclass classification.

Impact-slip experiments and systematic study of coal gangue “category” recognition technology part II: Improving effect of the proposed parallel voting system method on coal gangue “category” recognition accuracy based on impact-slip experiments

The research on coal gangue “category” recognition via single classification algorithm for single-channel signal data shows that the classification and recognition capabilities of different algorithms on different channel signals and data types exhibit significant differences. In order to improve coal gangue classification and recognition effect, a coal gangue classification and recognition method of “multi-information fusion” based on the “parallel voting system (PVS)” was proposed in this study. The joint judgment of coal gangue “category” was completed through the combination of multiple algorithms or channel signal data with stronger recognition ability according to the voting principle of “the minority is subordinate to the majority”. Based on the signal characteristics data obtained from the impact-slip test between coal gangue and the hydraulic support, three different modes of PVS-based coal gangue classification and recognition, namely, channel signal data PVS (SDPVS), algorithm PVS (APVS), and algorithm and channel signal data combination double PVS (algorithm-data double parallel, A&SDPVS), were respectively investigated herein. The influence of signals combination mode, algorithms matching mode as well as combined mode of signals and algorithms on coal gangue “category” voting recognition accuracy was studied. The research results show that for the SDPVS when a single algorithm with higher recognition accuracy and its corresponding multiple channel signal data with higher recognition accuracy were selected and for the A&SDPVS when the multiple “Recognition combination of single algorithm and single channel signal data” with higher recognition accuracy was selected, coal gangue classification and recognition effect could be improved. PVS-based coal gangue “category” recognition accuracy could reach up to 97.5%, which effectively improved coal gangue classification and recognition ability; moreover, the effectiveness of the multi-information fusion method proposed in this study was proven.

Metaheuristic Optimized Multi-Level Classification Learning System for Engineering Management

Multi-class classification is one of the major challenges in machine learning and an ongoing research issue. Classification algorithms are generally binary, but they must be extended to multi-class problems for real-world application. Multi-class classification is more complex than binary classification. In binary classification, only the decision boundaries of one class are to be known, whereas in multiclass classification, several boundaries are involved. The objective of this investigation is to propose a metaheuristic, optimized, multi-level classification learning system for forecasting in civil and construction engineering. The proposed system integrates the firefly algorithm (FA), metaheuristic intelligence, decomposition approaches, the one-against-one (OAO) method, and the least squares support vector machine (LSSVM). The enhanced FA automatically fine-tunes the hyperparameters of the LSSVM to construct an optimized LSSVM classification model. Ten benchmark functions are used to evaluate the performance of the enhanced optimization algorithm. Two binary-class datasets related to geotechnical engineering, concerning seismic bumps and soil liquefaction, are then used to clarify the application of the proposed system to binary problems. Further, this investigation uses multi-class cases in civil engineering and construction management to verify the effectiveness of the model in the diagnosis of faults in steel plates, quality of water in a reservoir, and determining urban land cover. The results reveal that the system predicts faults in steel plates with an accuracy of 91.085%, the quality of water in a reservoir with an accuracy of 93.650%, and urban land cover with an accuracy of 87.274%. To demonstrate the effectiveness of the proposed system, its predictive accuracy is compared with that of a non-optimized baseline model, single multi-class classification algorithms (sequential minimal optimization (SMO), the Multiclass Classifier, the Naïve Bayes, the library support vector machine (LibSVM) and logistic regression) and prior studies. The analytical results show that the proposed system is promising project analytics software to help decision makers solve multi-level classification problems in engineering applications.

Metagenomic Geolocation Prediction Using an Adaptive Ensemble Classifier.

Microbiome samples harvested from urban environments can be informative in predicting the geographic location of unknown samples. The idea that different cities may have geographically disparate microbial signatures can be utilized to predict the geographical location based on city-specific microbiome samples. We implemented this idea first; by utilizing standard bioinformatics procedures to pre-process the raw metagenomics samples provided by the CAMDA organizers. We trained several component classifiers and a robust ensemble classifier with data generated from taxonomy-dependent and taxonomy-free approaches. Also, we implemented class weighting and an optimal oversampling technique to overcome the class imbalance in the primary data. In each instance, we observed that the component classifiers performed differently, whereas the ensemble classifier consistently yielded optimal performance. Finally, we predicted the source cities of mystery samples provided by the organizers. Our results highlight the unreliability of restricting the classification of metagenomic samples to source origins to a single classification algorithm. By combining several component classifiers via the ensemble approach, we obtained classification results that were as good as the best-performing component classifier.

Genetic Programming With a New Representation to Automatically Learn Features and Evolve Ensembles for Image Classification.

Image classification is a popular task in machine learning and computer vision, but it is very challenging due to high variation crossing images. Using ensemble methods for solving image classification can achieve higher classification performance than using a single classification algorithm. However, to obtain a good ensemble, the component (base) classifiers in an ensemble should be accurate and diverse. To solve image classification effectively, feature extraction is necessary to transform raw pixels into high-level informative features. However, this process often requires domain knowledge. This article proposes an evolutionary approach based on genetic programming to automatically and simultaneously learn informative features and evolve effective ensembles for image classification. The new approach takes raw images as inputs and returns predictions of class labels based on the evolved classifiers. To achieve this, a new individual representation, a new function set, and a new terminal set are developed to allow the new approach to effectively find the best solution. More important, the solutions of the new approach can extract informative features from raw images and can automatically address the diversity issue of the ensembles. In addition, the new approach can automatically select and optimize the parameters for the classification algorithms in the ensemble. The performance of the new approach is examined on 13 different image classification datasets of varying difficulty and compared with a large number of effective methods. The results show that the new approach achieves better classification accuracy on most datasets than the competitive methods. Further analysis demonstrates that the new approach can evolve solutions with high accuracy and diversity.

Academic Performance Prediction Method of Online Education using Random Forest Algorithm and Artificial Intelligence Methods

In order to improve the teaching quality of online education, the prediction method of students' online academic performance has been studied. First, the learning analysis, artificial intelligence (AI) and other related theoretical concepts are analyzed and introduced. Then, the decision tree of single classification algorithm and the random forest (RF) of ensemble learning algorithm are analyzed, and the academic performance prediction model of online education is constructed by RF algorithm. Finally, the data of education platform is used for empirical analysis to verify the reliability and practicability of the academic performance prediction algorithm of online education. The connotation of learning analysis, the role and elements of learning analysis in the learning process are introduced. The algorithm principle of RF and decision tree is analyzed. By using the idea of information entropy and discretization, the continuous variables are processed to improve the fitting degree of the algorithm. The model is evaluated by empirical analysis, and the test accuracy of several different algorithms is compared. It is found that the prediction accuracy of the RF algorithm is more than 90%, which shows that the prediction method can help teachers and students to carry out better teaching and learning activities, so as to better improve students' ability to master knowledge. It is hoped that the result can provide some reference for the management of students' learning behavior and the optimization of teachers' teaching strategies in online learning activities

Multicategory Angle-Based Learning for Estimating Optimal Dynamic Treatment Regimes With Censored Data

An optimal dynamic treatment regime (DTR) consists of a sequence of decision rules in maximizing long-term benefits, which is applicable for chronic diseases such as HIV infection or cancer. In this article, we develop a novel angle-based approach to search the optimal DTR under a multicategory treatment framework for survival data. The proposed method targets to maximize the conditional survival function of patients following a DTR. In contrast to most existing approaches which are designed to maximize the expected survival time under a binary treatment framework, the proposed method solves the multicategory treatment problem given multiple stages for censored data. Specifically, the proposed method obtains the optimal DTR via integrating estimations of decision rules at multiple stages into a single multicategory classification algorithm without imposing additional constraints, which is also more computationally efficient and robust. In theory, we establish Fisher consistency and provide the risk bound for the proposed estimator under regularity conditions. Our numerical studies show that the proposed method outperforms competing methods in terms of maximizing the conditional survival probability. We apply the proposed method to two real datasets: Framingham heart study data and acquired immunodeficiency syndrome clinical data. Supplementary materials for this article are available online.

Detection of Intruder in Cloud Computing Environment using Swarm Inspired based Neural Network

Cloud computing services offered a resource pool with a wide range of storage for large amounts of data. Cloud services are generally used as a demand-driven private or open data forum, and the increase in use has led to security concerns. Therefore, it is necessary to design an accurate Intrusion Detection System (IDS) to identify the suspected node in the cloud computing environment. This is possible by monitoring network traffic so that the quality of service and performance of the system can be maintained. Several researchers have worked on designing valid IDS with the help of a machine learning approach. A single classification algorithm seems to be impossible to detect intruders with high accuracy. Therefore, a hybrid approach is presented. This approach is a combination of Cuckoo Search. CS as an optimization algorithm and Feed Forward Back Propagation Neural Network (FFBPNN) as a multi-class classification approach. The user's request to access cloud data is collected and essential features are selected using CS as an optimization approach. The selected features are used to train FFBPNN with reduced training time and complexity. The experimental analysis has been performed in terms of precision, recall, F-measure, and accuracy. The evaluated value for parameters i.e., precision (85.5%), recall (86.4%), F-measure (85.9%), and accuracy (86.22%) are observed. At last, the parameters are also compared with the existing approach.

Research on Emotion Classification based on Complex Network and Ensemble Learning

Because the traditional feature extraction is based on the statistical information such as document frequency and word frequency, the selection of feature words is ignored, and the semantic correlation between words in the text is ignored. The feature selection method based on complex network takes into account the semantic association between words, but does not take into account the statistical information such as word frequency. The above methods are not satisfactory for the selection of feature words, which affects the effect of text classification. Therefore, this paper combines the two, proposes a new method for feature selection, and in order to solve the problem of low accuracy rate of single classification algorithm, USES integrated learning [1] to strengthen the classification algorithm. The results show that this method is feasible and achieves good classification effect.