检索结果-内蒙古大学图书馆

A novel SSA-CatBoost machine learning model for credit rating

JOURNAL OF INTELLIGENT & FUZZY SYSTEMS 2023年第2期44卷 2269-2284页

作者： Yang, Ruicheng Wang, Pucong Qi, Ji Inner Mongolia Univ Finance & Econ Finance Sch Hohhot 010000 Inner Mongolia Peoples R China

Categorical Boost (CatBoost) is a new approach in credit rating. In the process of classification and prediction using CatBoost, parameter tuning and feature selection are two crucial parts, which affect the classification accuracy of CatBoost significantly. This paper proposes a novel SSA-CatBoost model, which mixes sparrow search algorithm (SSA) and CatBoost to improve classification and prediction accuracy for credit rating. In terms of parameter tuning, the SSA-CatBoost optimization obtains the most optimal parameters by iterating and updating the sparrow's position, and utilize the optimal parameter to improve the accuracy of classification and prediction. In terms of feature selection, a novel wrapping method called Recursive Feature Elimination algorithm is adopted to reduce the adverse impact of noise data on the results, and further improves calculation efficiency. To evaluate the performance of the proposed SSA-CatBoost model, P2P lending datasets are employed to assess the prediction results, then the interpretable Shap package is used to explain the reason why the proposed model considers a sample as good or bad. Consequently, the experimental results show that the SSA-CatBoost model has an ideal accuracy in classification and prediction for credit rating by comparing the SSA-CatBoost model with the CatBoost model and other well-known machine learning models.

关键词： CatBoost sparrow search algorithm parameter tuning feature selection credit rating

来源：评论

学校读者我要写书评

暂无评论

A novel strategy for efficient biodiesel production: Optimization, prediction, and mechanism

引用

RENEWABLE ENERGY 2023年 207卷 385-397页

作者： Wang, Xiao-Man Zeng, Ya-Nan Wang, Yu-Ran Wang, Fu-Ping Wang, Yi-Tong Li, Jun-Guo Ji, Rui Kang, Le-Le Yu, Qing Liu, Tian-Ji Fang, Zhen North China Univ Sci & Technol Coll Met & Energy 21 Bohai St Tangshan 063210 Peoples R China Southern Med Univ Biomat Res Ctr Sch Biomed Engn Guangdong Prov Key Lab Construction & Detect Tissu Guangzhou 510515 Peoples R China Nanjing Agr Univ Coll Engn Biomass Grp 40 Dianjiangtai Rd Nanjing 210031 Peoples R China

Resistance of biodiesel industrial production came from high energy consumption and feedstock costs. To solve it, Na2CO3@BFD catalyst was prepared from blast furnace dust and used to catalyze biodiesel production at low temperature. Biodiesel yield of 99.04 wt% was obtained under conditions optimized by response surface methodology of methanol/oil molar ratio 13.72/1, catalyst dosage 9.77 wt % and 74.86 degrees C for 1.62 h. The order of influence of the four factors was temperature (245.9) > time (109.8) > methanol/oil molar ratio (23.83) > catalyst dosage (1.19). Back propagation neural network model (BPNN) was optimized using genetic algorithm (GA) and sparrow search algorithm (SSA) to predict biodiesel yield. The evaluation indexes of mean absolute error, mean square error, root mean square error and mean absolute percentage error of SSA-BPNN were 0.9236, 2.0184, 1.4207 and 1.0247 (vs. 2.4329, 9.1037, 3.0172 and 3.5000 for GA-BPNN and 4.3291, 43.4693, 6.5931 and 6.9227 for BPNN), indicating that SSA-BPNN model had excellent prediction ability to effectively reduce experimental costs and resource consumption. The reaction kinetics of Na2CO3@BFD for transesterification process showed that its activation energy was 65.73 kJ/mol, lower than that of reported solid base catalyst, indicating that it had significant potential in biomass conversion.

关键词： Biodiesel Response surface methodology Back propagation neural network Genetic algorithm sparrow search algorithm Kinetics

来源：评论

学校读者我要写书评

暂无评论

Interactive Park Landscape Environment Design by Incorporating Improved SOM algorithm

引用

JOURNAL OF ADVANCED COMPUTATIONAL INTELLIGENCE AND INTELLIGENT INFORMATICS 2025年第2期29卷 241-255页

作者： Cui, Bingsheng Wang, Shuai Qiqihar Univ Coll Arts & Design 35 Zhonghua West Rd Qiqihar 161000 Heilongjiang Peoples R China Hulunbuir Univ Acad Fine Arts 26 Cent Genghis Khan Rd Hulunbuir 021000 Inner Mongolia Peoples R China

In view of the inadequacy of interactivity and ecological aesthetics in park landscape, the existing methods are often difficult to achieve accurate terrain analysis and design, resulting in unsatisfactory landscape effects. In order to make up for this shortcoming, the research innovatively integrates sparrow search algorithm (SSA) and self-organizing mapping algorithm (SOM) to conduct in-depth analysis of park landscape topography. SSA, with its powerful global search ability, can quickly lock the optimal solution region. SOM uses unsupervised learning to effectively mine the underlying rules in terrain data. The combination of the two not only improves the accuracy and efficiency of terrain analysis, but also lays a solid foundation for the construction of park landscape environment interaction design model. The root-mean-square error of the algorithm is only 0.01 m2, the accuracy is as high as 98.1%, and the F1-value is also as high as 97.8%. Compared with previous studies, the algorithm has better fitting performance. In addition, the model is applied to the actual park landscape design, and an interactive park landscape optimization design scheme is proposed based on the design concept of forest walk and ecological protection area. After evaluation, the program scored 9.1 points for interactivity and 9.2 points for sustainability, fully proving that it meets the interactive needs of tourists while also taking into account ecological protection and sustainability. Compared with previous studies, the innovation of this study lies in the successful integration of SSA and SOM algorithms, which significantly improves the accuracy of terrain analysis, and verifies the practicability and effectiveness of this method through practical application. Overall, the method proposed in the study can analyze topography more accurately than traditional methods and design park landscapes that are both interactive and aesthetically pleasing.

关键词： sparrow search algorithm self-organizing mapping algorithm interactive park landscape environment design ecological conservation

来源：评论

学校读者我要写书评

暂无评论

Optimal Design of MPC Autonomous Vehicle Trajectory Tracking Controller Considering Variable Time Domain

引用

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING 2025年第8期50卷 5697-5710页

作者： Ma, Hao Pei, Wenhui Zhang, Qi Shandong Jiaotong Univ Sch Informat Sci & Elect Engn Jinan 250357 Peoples R China Shandong Univ Sch Control Sci & Engn Jinan 250061 Peoples R China

In recent years, with the indepth research on driverless technology, model predictive control theory was extensively applied in the field of vehicle control. In order to improve the accurate tracking of reference trajectories by driverless vehicles, a model predictive control trajectory tracking controller for driverless vehicles optimized by an improved sparrow search algorithm is proposed. Firstly, an objective function with constraints is added to the model predictive control trajectory tracking controller by establishing the vehicle dynamics model;Secondly, the improved sparrow search algorithm is enhanced to speed up convergence and expand the program's search capabilities;Then, in order to discover the best value, the model predictive control trajectory tracking controller's prediction time domain and control time domain are optimized using the improved sparrow search algorithm;Finally, to confirm the method's viability, collaborative simulations in Simulink/Carsim were completed. The simulation results show that the lateral errors generated by the improved sparrow search algorithm-based optimized model predictive control trajectory tracking controller are reduced by 53.53% and 65.44%, respectively, when the vehicle speed is 36 km/h, compared with the traditional model predictive control trajectory tracking controller. When the vehicle speed is 54 km/h, the lateral deviations are reduced by 81.08% and 86.76%, respectively. In addition, the optimized model predictive control trajectory tracking controller improves the accuracy and at the same time, the driving stability of the control vehicle is significantly improved.

关键词： Driverless sparrow search algorithm Model predictive control Trajectory tracking

来源：评论

学校读者我要写书评

暂无评论

Rapid stability assessment of barrier dams based on the extreme gradient boosting model

引用

NATURAL HAZARDS 2025年第3期121卷 3047-3072页

作者： Yang, Haiqing Li, Hao Chen, Chiwei Liu, Xinchang Chongqing Univ Sch Civil Engn Key Lab New Technol Construct Cities Mt Area Chongqing 400045 Peoples R China Natl Joint Engn Res Ctr Geohazards Prevent Reservo Chongqing 400045 Peoples R China

Barrier dams are major natural disasters that frequently occur in mountainous areas and have a high probability of destabilizing and failing within a short period. Catastrophic dam failures cause significant economic and ecological damage to downstream areas. Therefore, rapid and accurate assessment of barrier dam stability is crucial for effective emergency rescue efforts. A database was constructed with 1,738 barrier dam cases from 49 countries and regions. A machine learning algorithm was employed for barrier dam stability assessment, and the sparrow search algorithm (SSA) was used to optimize the Gradient Boosting Decision Tree, Random forest, and Extreme Gradient Boosting (XGBoost) respectively. Then, 192 cases with detailed data were selected for training, validation, and comparison of model performance. Among the methods tested, the SSA-XGBoost model performed the best. In addition, the feature importance values of the input parameters were also analyzed. The dam height emerged as the most important factor, with a significantly higher importance value compared to other features. Based on this, 50 barrier dam cases with detailed information were randomly selected for stability assessment, and the results were compared with those of several classical methods. The results indicate that the SSA-XGBoost algorithm can more accurately judge the stability of barrier dams. The absolute accuracy, conservative accuracy, and misjudgment rates were 86%, 94%, and 6%, respectively. These findings demonstrate that the assessment method proposed in this study is reliable and can provide some practical references for the emergency rescue of barrier dams.

关键词： Barrier dam stability Rapid assessment method sparrow search algorithm Extreme gradient boosting

来源：评论

学校读者我要写书评

暂无评论

Dynamic Response Evaluation for Single-Hole Bench Carbon Dioxide Blasting Based on the Novel SSA-VMD-PCC Method

引用

INTERNATIONAL JOURNAL OF GEOMECHANICS 2023年第1期23卷 p1-12页

作者： Zhou, Shengtao Yao, Yingkang Luo, Xuedong Jiang, Nan Niu, Shuaishuai Jianghan Univ State Key Lab Precis Blasting Hubei Key Lab Blasting Engn Wuhan 430056 Peoples R China China Univ Geosci Fac Engn Wuhan 430074 Peoples R China

Carbon dioxide blasting is an excellent alternative technique for blasting. Nowadays, bench carbon dioxide blasting has been gradually used in mining and rock engineering. To evaluate the essential characteristics of the ground vibration induced by bench carbon dioxide blasting, we monitored the single-hole bench carbon dioxide blasting vibration signal and proposed a novel method for determining variational modal decomposition (VMD) parameters based on the sparrow search algorithm (SSA). After the signal has been decomposed by the VMD method with the optimized initial parameters, combined with Pearson correlation coefficients (PCC), the noise and false components from the original signal can be removed. Then, the time-frequency characteristics of the monitored ground vibration were analyzed through the Hilbert transform. Latterly, based on the monitored peak particle velocity (PPV) in carbon dioxide blasting, an empirical equation for the ground vibration was established. It is indicated from the results that the SSA-VMD-PCC method can be successfully applied to the vibration signal preprocessing of carbon dioxide blasting. The vibration energy of bench carbon dioxide blasting is mainly concentrated in the range of 0-128 Hz. The attenuation law of PPVs in two monitoring directions obeys the power function, and the PPVs in the jet direction are greater than that in the vertical jet direction. Compared with the single free-surface blasting, ground vibration excited by the bench blasting with two free surfaces is weaker. Moreover, the blasting vibration empirical equation proposed by dimensional analysis can effectively describe the attenuation of the PPV in bench carbon dioxide blasting.

关键词： Carbon dioxide blasting Blasting vibration sparrow search algorithm Peak particle velocity Empirical equation

来源：评论

学校读者我要写书评

暂无评论

Geographical Origin Identification of Chinese Tomatoes Using Long-Wave Fourier-Transform Near-Infrared Spectroscopy Combined with Deep Learning Methods

引用

FOOD ANALYTICAL METHODS 2023年第4期16卷 664-676页

作者： Yuan, Weidong Jiang, Hongzhe Sun, Mengmeng Zhou, Yu Zhang, Cong Zhou, Hongping Nanjing Forestry Univ Coinnovat Ctr Efficient Proc & Utilizat Forest Res Nanjing 210037 Peoples R China Nanjing Forestry Univ Coll Mech & Elect Engn Nanjing 210037 Peoples R China

Tomato cultivation in China is concentrated in Xinjiang, Henan, and Shandong provinces, and the quality of tomatoes produced in different regions shows huge differences. Xinjiang tomatoes have extraordinary nutritional compositions and sensory qualities, and counterfeit Xinjiang tomato products are proliferating in the market. This study aimed to investigate the feasibility of identifying the geographical origin of Chinese tomatoes by using long-wave Fourier-Transform near-infrared spectroscopy (FT-NIR, 10,000-4000 cm(-1)). First, principal component analysis (PCA) was conducted on the raw spectra, and it was found that the first two PCs effectively identified the geographical origin of the tomatoes. Meanwhile, the results of partial least squares discriminant analysis (PLS-DA) combined with different preprocessing methods showed that the PLS-DA model based on the raw spectra achieved the best performance. The optimal PLS-DA model achieved the correct classification rate (CCR) of 97.8% in an external prediction set, and showed that raw spectra contained sufficient valid information. Besides, six algorithms, grid search (GS), genetic algorithm (GA), particle swarm algorithm (PSO), grey wolf algorithm (GWO), improved grey wolf algorithm (IGWO), and sparrow search algorithm (SSA) were employed to optimize the parameters of support vector machine (SVM). The SSA-SVM model exhibited the best performance with a CCR of 97.8% in the prediction set. Afterward, 40 and 8 spectral features were extracted from the raw full spectra using stacked autoencoder (SAE) and PC loading, respectively. Finally, the qualitative analysis model based on the feature variables was further investigated, and the SAE-SSA-SVM simplified model showed the best performance with a CCR of 98.7%, 95.6%, and 95.6% in the calibration, cross-validation, and prediction set, respectively. The results of this study provide theoretical support for applying long-wave FT-NIRS combined with machine learning and dee

关键词： Fourier-Transform near-infrared spectroscopy Tomatoes Origin identification sparrow search algorithm Stacked auto-encoders

来源：评论

学校读者我要写书评

暂无评论

Regionalized fault line in distribution networks based on an improved SSA-VMD and multi-scale fuzzy entropy

引用

ELECTRICAL ENGINEERING 2023年第6期105卷 4399页

作者： Chen, Bofan Sun, Yanzhou Song, Xiaoyan Wang, Bin Henan Polytech Univ Sch Elect Engn & Automat Jiaozuo 454000 Peoples R China Henan Elect Power Co Elect Power Res Inst Zhengzhou 450052 Peoples R China

To solve the insufficient utilization of transient power frequency current components when single-phase grounding fault occurs in high-voltage grounding system, a ground fault routing method based on improved sparrow search algorithm optimized variational modal decomposition (ISSA-VMD) and multi-scale fuzzy entropy (MFE) is proposed. First, the elite opposition-based learning strategy (EOBLS) is used to improve the population diversity of the SSA and adopt the improved SSA to optimize VMD's parameters. From the experimental data, we can know the optimized variational mode decomposition can accurately distinguish the power frequency components of each feeder transient zero-sequence current. Second, calculate the MFE value of the power frequency component of the zero-sequence current of each feeder and select the multi-scale fuzzy entropy partial mean (PMMFE) as the line selection criterion, and then, the fault line is selected. MATLAB/Simulink simulation experimental results and the experimental results of a real experimental field for a 10-kV distribution network show that this proposed method can select the right line under most mistake conditions, with high reliability and strong robustness.

关键词： Fault line selection Elite opposition-based learning strategy sparrow search algorithm Variational modal decomposition Multi-scale fuzzy entropy

来源：评论

学校读者我要写书评

暂无评论

Recognizing defects in stainless steel welds based on multi-domain feature expression and self-optimization

引用

JOURNAL OF INTELLIGENT MANUFACTURING 2023年第3期34卷 1293-1309页

作者： Zhang, Rui Zhao, Na Fu, Liuhu Bai, Xiaolu Cai, Jianghui Taiyuan Univ Sci & Technol Shanxi Key Lab Adv Control & Equipment Intelligen Taiyuan 030024 Peoples R China Taiyuan Univ Sci & Technol Coll Comp Sci & Technol Taiyuan 030024 Peoples R China Shanxi Design & Res Inst Mech & Elect Engn Co Ltd Taiyuan 030009 Peoples R China

The recognition of different welding defects is important for the assessment of the safety of welded structures. This study proposes a method to recognize defects in stainless steel welds based on multi-domain feature expression and self-optimization. This is because of the poor detection of feature-related information from signals, inadequate feature extraction by the convolutional network, the limited capability of intelligent techniques, network redundancy, and a lack of self-adaptive capability in prevalent methods, A 1D ultrasonic detection dataset of austenitic stainless steel welds in the time domain (TD) is first constructed and the 1D TD signals are rendered in the time-frequency domain, Gramain angular field, and the Markov transition field. The aim is to enrich the feature expression of 1D ultrasonic echo data of the weld defects. A comparison among a variety of lightweight convolutional neural networks on multi-spatial domain datasets is used to identify a combination of network and spatial domain datasets that are suitable for recognizing welding defects. Finally, a multi-scale depthwise separable convolution is designed, and is subjected to adaptive compression and parameter-adaptive optimization based on the sparrow search algorithm to construct the self-optimizing lightweight multi-scale MobileNetV3 (SLM-MobileNetV3) model. The results of experiments showed that the SLM-MobileNetV3 model has an accuracy of 97.26% for the recognition of five types of defects: inclusion, crack, porosity, incomplete penetration, and a lack of fusion. The time required for testing a single image was 2 ms. Experimental analysis showed that the proposed method can improve the accuracy of defect recognition while using few parameters and computational cost, and is generalizable.

关键词： Welding defect Ultrasonic detection Multi-domain and multi-scale analysis sparrow search algorithm Self-optimization

来源：评论

学校读者我要写书评

暂无评论

CSSA-based collaborative optimization recommendation of users in mobile crowdsensing

引用

PEER-TO-PEER NETWORKING AND APPLICATIONS 2023年第2期16卷 803-817页

作者： Wang, Jian Hao, Shuai Zhao, Guosheng Harbin Univ Sci & Technol Sch Comp Sci & Technol Harbin 150080 Peoples R China Harbin Normal Univ Coll Comp Sci & Informat Engn Harbin 150025 Peoples R China

With the large-scale popularization of mobile terminals, crowd sensing technology has gradually replaced the existing static sensors with its efficient and low-cost advantages as an emerging data collection method. How to quickly allocate the sensing task to the optimal execution user under the premise of ensuring the perceived quality and reducing the cost is the focus of the research. In this regard, this paper proposes a Crowd sensing sparrow search algorithm (CSSA) collaborative optimization recommendation method that combines fitness priority, collaboration, and intelligent optimization algorithms, and uses it for task allocation problems. Firstly, the concept of fitness is proposed to calculate the location, power, equipment and reputation of the perceived user, and analyze the matching degree of the user to the task. Secondly, according to the different fitness, the user is divided into explorers and followers, and the two cooperate to complete the perception task. Thirdly, in the process of solving the optimal task allocation scheme, CSSA intelligent optimization algorithm is used to simulate the process of users completing tasks, and the selected user results can be obtained after limited iterations. Through the comparative experiments of the proposed algorithm and other optimization algorithms in the same environment, the results show that it has higher performance in solving the task allocation problem.

关键词： Mobile crowdsensing sparrow search algorithm Task allocation User collaboration Priority

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：