检索结果-内蒙古大学图书馆

Multi-response Optimization of 3D Printed Parts with Triangular Patterns Using Nonlinear machine learning Regressor Technique

引用

JOURNAL OF MATERIALS ENGINEERING AND PERFORMANCE 2024年 1-17页

作者： Tandon, Sourabh Singh, Sanjay Kumar Kacker, Ruchin Gautam, Satyam Shivam Tamang, Santosh Kumar Indian Inst Technol Bhubaneswar Bhubaneswar Orissa India Sagar Inst Sci & Technol Dept Mech Engn Bhopal Madhya Pradesh India NERIST Dept Mech Engn Itanagar Arunachal Prade India

In the ever-changing field of additive manufacturing, this work implemented machine learning algorithm (nonlinear regression approach) to find out the optimal tensile strength and modulus of elasticity of 3D-printed ABS, PLA, and PLA + CF components using MATLAB R2021a. The work navigates the complex interactions among material type, orientation, and infill density using a complex experimental design matrix. With an impressive accuracy rate of more than 90%, the nonlinear model proves to be a useful tool for predicting mechanical properties in the context of 3D printing. The study uses the desirability technique as part of a multi-response optimization strategy to achieve optimal performance. By using this procedure, the best process parameters are identified, and the result is a composite desirability score of 0.84. An orientation fixed at 0 degrees, a material type defined as PLA + CF, and an infill density set at 80% comprise the identified peak of efficiency. In addition, the ideal parameters for the responses of Elasticity Modulus and Tensile Strength are 80% infill density, 90 degrees orientation, and PLA material type. These results not only identify the best possible configurations, but also offer a detailed insight into the complex relationship between printing settings and the mechanical qualities that are produced. A Validation tests is carried out to demonstrate that the model is reliable and has a predicted accuracy of more than 90%. This strong validation demonstrates how well the model works in real-world scenarios and gives rise to confidence in its capacity to direct additive manufacturing procedures toward improved mechanical performance. This work essentially illuminates the intricacies of 3D printing parameters and provides a clear path for maximizing mechanical qualities, making it a lighthouse in the field of additive manufacturing. The research findings offer significant assistance for engineers and designers who aim to optimize their processe

关键词： additive manufacturing machine learning algorithm mechanical properties optimization

来源：评论

学校读者我要写书评

暂无评论

machine learning approach for the effectual production of a novel esterase and its application in bioremediation of dairy effluent

引用

PROCESS BIOCHEMISTRY 2024年 144卷 20-37页

作者： Sounderarajan, Sathieesh Karuppusamy, Kanakalakshmi Surya, Aafreen Periyar Vikram Puchalapalli, Dinesh Sankar Reddy Sethi, Gautam Ayothiraman, Seenivasan Natl Inst Technol Andhra Pradesh Dept Chem Engn Tadepalligudem 534101 Andhra Pradesh India Natl Inst Technol Andhra Pradesh Dept Biotechnol Biochem Engn Res Grp Tadepalligudem 534101 Andhra Pradesh India Indian Inst Technol Madras Dept Chem Engn Chennai 600036 Tamil Nadu India Bharathidasan Univ Dept Biotechnol Tiruchirappalli 620024 Tamil Nadu India Natl Univ Singapore NUS Ctr Canc Res Yong Loo Lin Sch Med Dept Pharmacol Singapore Singapore

The present work intends to maximize the production of an industrially important novel esterase from Trichoderma harzianum using both machine learning and statistical approaches and its application in the bioremediation of dairy effluent. Significant medium compounds, such as almond oil, casein, NaH 2 PO 4 , and trace elements, were screened from a pool of carbon, nitrogen, oil and surfactant sources along with other micronutrients using a first order statistical design. The optimum value of screened parameters was found using response surface methodology that resulted in the maximum esterase production of 2985.31 U/L. Artificial neural network trained with the experimental response of RSM experiments coupled with genetic algorithm optimization technique has resulted in the maximum esterase production of 3256.73 U/L, which is 150-folds higher as compared to the unoptimized condition and also found to be the higher than the outcome of statistical optimization. Dairy effluent was used for the production of esterase as well as for its bioremediation. At 50% load, the maximum reduction in chemical oxygen demand and the production of esterase were observed to be 46.84% and 34.12 U/ L, respectively. The COD reduction of diary effluent was modelled using Michaelis-Menten kinetics and the kinetic parameters, half saturation constant (K m ) and maximum reaction rate (r max ) value were evaluated as 286.88 mg/L and 26.04 mg/L/d, respectively.

关键词： Esterase Trichoderma harzianum Optimization machine learning algorithm Dairy effluent treatment

来源：评论

学校读者我要写书评

暂无评论

The Utility Impact of Differential Privacy on Credit Card Data in machine learning algorithms

引用

Procedia Computer Science 2023年 221卷 664-672页

作者： Xiaopeng LUO Siyuan WANG Haolong CHEN Zongwei LUO BNU-UIC Institute of AI and Future Networks Beijing Normal University Zhuhai China 519000

With the development of networks and financial technologies, credit card data is increasingly being used in various fields of data analysis such as user behavior, financial transactions, and market analysis. These fields often use multiple machine learning algorithms for data mining on credit card dataset. It is worth noting that credit card data contains more diverse and comprehensive information compared to traditional data, and the dataset may contain multiple data types. At the same time, credit card data may also expose users’ privacy information. Differential privacy algorithms can add random noise to the data set, protecting sensitive information while ensuring certain data utility. However, there has been little research on the use of differential privacy algorithms on credit card data in multiple machine learning algorithms, and there has been insufficient exploration of the utility impact of differential privacy on complex credit card data. These research gaps exist in both the financial technology and privacy protection industries. Therefore, this paper applies differential privacy to credit card data in multiple classic machine learning algorithms, discusses the utility impact of various differential privacy algorithms on credit card data, and compares the performance of credit card data sets protected by differential privacy in different algorithms.

关键词： credit card default differential privacy machine learning algorithm

来源：评论

学校读者我要写书评

暂无评论

machine learning algorithms for Chromophoric Dissolved Organic Matter (CDOM) Estimation Based on Landsat 8 Images

引用

REMOTE SENSING 2021年第18期13卷 3560-3560页

作者： Sun, Xiao Zhang, Yunlin Zhang, Yibo Shi, Kun Zhou, Yongqiang Li, Na Chinese Acad Sci State Key Lab Lake Sci & Environm Nanjing Inst Geog & Limnol Nanjing 210008 Peoples R China

Chromophoric dissolved organic matter (CDOM) is crucial in the biogeochemical cycle and carbon cycle of aquatic environments. However, in inland waters, remotely sensed estimates of CDOM remain challenging due to the low optical signal of CDOM and complex optical conditions. Therefore, developing efficient, practical and robust models to estimate CDOM absorption coefficient in inland waters is essential for successful water environment monitoring and management. We examined and improved different machine learning algorithms using extensive CDOM measurements and Landsat 8 images covering different trophic states to develop the robust CDOM estimation model. The algorithms were evaluated via 111 Landsat 8 images and 1708 field measurements covering CDOM light absorption coefficient a(254) from 2.64 to 34.04 m(-1). Overall, the four machine learning algorithms achieved more than 70% accuracy for CDOM absorption coefficient estimation. Based on model training, validation and the application on Landsat 8 OLI images, we found that the Gaussian process regression (GPR) had higher stability and estimation accuracy (R-2 = 0.74, mean relative error (MRE) = 22.2%) than the other models. The estimation accuracy and MRE were R-2 = 0.75 and MRE = 22.5% for backpropagation (BP) neural network, R-2 = 0.71 and MRE = 24.4% for random forest regression (RFR) and R-2 = 0.71 and MRE = 24.4% for support vector regression (SVR). In contrast, the best three empirical models had estimation accuracies of R-2 less than 0.56. The model accuracies applied to Landsat images of Lake Qiandaohu (oligo-mesotrophic state) were better than those of Lake Taihu (eutrophic state) because of the more complex optical conditions in eutrophic lakes. Therefore, machine learning algorithms have great potential for CDOM monitoring in inland waters based on large datasets. Our study demonstrates that machine learning algorithms are available to map CDOM spatial-temporal patterns in inland waters.

关键词： machine learning algorithm support vector regression (SVR) Gaussian process regression (GPR) chromophoric dissolved organic matter (CDOM) Landsat 8

来源：评论

学校读者我要写书评

暂无评论

machine learning Regression for Assessing Sensing Performance and Anticancer Potential of Oolong Tea-Derived Silver Nanoparticles

引用

JOURNAL OF THE ELECTROCHEMICAL SOCIETY 2025年第3期172卷 037503-037503页

作者： Sert, Esra Kalindemirtas, Ferdane Danisman Karakus, Emir Ersel Erol, Ayse Ozbas, Fatih Kucukdeniz, Tarik Karakus, Selcan Istanbul Univ Istanbul Fac Med Dept Hematol TR-34390 Istanbul Turkiye Erzincan Binali Yildirim Univ Fac Med Dept Physiol Erzincan Turkiye Private Nazmi Arikan Sci Sch TR-34906 Istanbul Turkiye Istanbul Univ Istanbul Fac Med Dept Med Biol Istanbul Turkiye Fatih Sultan Mehmet Vakif Univ Fac Arts Design & Architecture Dept Restorat & Conservat Cultural Assets TR-34015 Istanbul Turkiye Istanbul Univ Cerrahpasa Fac Engn Dept Ind Engn TR-34320 Istanbul Turkiye Istanbul Univ Cerrahpasa Fac Engn Dept Chem TR-34320 Istanbul Turkiye Hlth Biotechnol Joint Res & Applicat Ctr Excellenc TR-34220 Istanbul Turkiye

In this study, machine learning (ML) algorithms were employed to predict analyte concentrations using sensing results and evaluate the anticancer effects of nanostructures. Multifunctional oolong tea extract-mediated silver nanoparticles (OTE-Ag NPs) were synthesized via a photo/ultrasound method and utilized in various applications, including a smartphone-based H2O2 sensor and electrochemical sensors for urea and fructose. Key features were extracted from electrochemical results, and feature importance analysis was used to select the most predictive features. The artificial neural network (ANN) model provided accurate predictions, particularly strong for urea (R2 = 0.8575, RMSE = 0.4266, MAE = 0.3380). The study revealed the selective toxicity of OTE-Ag NPs to MCF-7 breast cancer cells through analyses of cytotoxicity, apoptosis, cell cycle phases, and CD44 surface marker expression using Annexin V/PI dye and flow cytometry. Experimental results demonstrated that OTE-Ag NPs suppressed MCF-7 cell proliferation while exhibiting lower cytotoxicity in normal HUVEC cells (46% cell death). OTE-Ag NPs arrested MCF-7 cells in the G2/M phase, induced apoptosis, and reduced CD44 expression, suggesting metastasis suppression. The CD44+/CD24- ratio decreased from 84.79% in control MCF-7 cells to 47.7% in OTE-Ag NP-treated cells. Overall, OTE-Ag NPs significantly inhibited MCF-7 cell proliferation through the apoptotic pathway by regulating the cell cycle in the G2/M phase.

关键词： machine learning algorithm artificial neural network apoptosis silver nanoparticles breast cancer electrochemical sensor

来源：评论

学校读者我要写书评

暂无评论

Applications of machine learning algorithms for biological wastewater treatment: Updates and perspectives

引用

CLEAN TECHNOLOGIES AND ENVIRONMENTAL POLICY 2021年第1期23卷 127-143页

作者： Sundui, Batsuren Ramirez Calderon, Olga Alejandra Abdeldayem, Omar M. Lazaro-Gil, Jimena Rene, Eldon R. Sambuu, Uyanga Natl Taiwan Univ Grad Inst Environm Engn Taipei 10617 Taiwan IHE Delft Inst Water Educ Dept Environm Engn & Water Technol POB 3015 NL-2601 DA Delft Netherlands Natl Univ Mongolia Sch Engn & Appl Sci Dept Informat & Comp Sci Bldg 3Baga Toiruu 2POB 46A-600 Sukhbaatar 14200 Ulaanbaatar Mongolia

Biological wastewater treatment using algae-bacteria consortia for nutrient uptake and resource recovery is a 'paradigm shift' from the mainstream wastewater treatment process to mitigate pollution and promote circular economy. The symbiotic relationship between algae and bacteria is complex in open or closed biological wastewater treatment systems. In this regard, machine learning algorithms (MLAs) have found to be advantageous to predict the uncertain performances of the treatment processes. MLAs have shown satisfactory results for effective real-time monitoring, optimization, prediction of uncertainties and fault detection of complex environmental systems. By incorporating these algorithms with online sensors, the transient operating conditions during the treatment process including disruptions or failures due to leaking pipelines, malfunctioning of bioreactors, unexpected fluctuations of organic loadings, flow rate, and temperature can be forecasted efficiently. This paper reviews the state-of-the-art MLA approaches for the integrated operation of biological wastewater treatment systems combining algal biomass production and nutrient recovery from municipal wastewater. [GRAPHICS] .

关键词： Algae-bacteria consortia machine learning algorithm Biological wastewater treatment Process optimization Bioprocess modelling

来源：评论

学校读者我要写书评

暂无评论

Comparison and analysis of accuracy of various machine learning algorithms in abnormal state monitoring of internet of things devices

Comparison and analysis of accuracy of various machine learn...

引用

作者： Bolun Zhang Beijing University of Post and Telecommunication

With the development of Internet of Things technology,more and more devices are connected to the Internet,including not only traditional computers,mobile phones and other smart terminal devices,but also various sensor *** sensor devices can collect a variety of environmental information and physical quantities,such as temperature,humidity,air pressure,light intensity,vibration,*** data have the characteristics of real-time,scale and diversity,and need to be processed and analyzed by appropriate *** the basis of previous studies,this project summarized the application of various machine learning algorithms in device state detection,compared the differences of various machine learning algorithms in sensor device detection and made comparative analysis,calculated the evaluation parameters of MSE,RMSE,MAE,MAPE,R and other aspects of the machine learning regression *** the effects of various regression models for better monitoring and prediction of equipment *** the analysis of a large number of historical data,different equipment state models can be established,and these models can be used to monitor and predict the current equipment *** can effectively avoid production line downtime or other losses caused by equipment failures or *** the same time,through the in-depth analysis of historical data,we can find some potential problems and take corresponding measures to prevent *** project aims to summarize the application of various machine learning algorithms in device status detection,compare and contrast the differences of various machine learning algorithms in sensor device detection,realize efficient processing and analysis of sensor data,calculate MSE,RMSE,MAE,MAPE,R and other evaluation parameters,and evaluate and compare each *** provide more accurate,reliable and efficient equipment condition monitoring and forecasting services for enterprises and individuals.

关键词： machine learning algorithm Equipment Condition Monitoring algorithm Structure

来源：评论

学校读者我要写书评

暂无评论

machine learning modelling of structural response for different seismic signal characteristics: A parametric analysis

引用

APPLIED SOFT COMPUTING 2024年 164卷

作者： De Iuliis, M. Miceli, E. Castaldo, P. Politecn Torino Dept Struct Geotech & Bldg Engn DISEG Turin Italy

The present study investigates the best seismic parameters for modeling the dynamic response of various nonlinear structural systems by comparing different machine learning (ML) algorithms. A total of 400 synthetic excitations were generated and analyzed against 23 seismic parameters. These signals were used in a step-by-step numerical analysis to calculate the dynamic responses of 1000 single-degree-of-freedom (SDOF) systems with varying mechanical properties. The data obtained from these responses were processed using 20 ML algorithms, including linear regression, tree, support vector machine (SVM), boosted and bagged trees, and artificial neural network (ANN). Each ML algorithm used a single seismic parameter as input to determine the most predictive parameters for modeling structural responses, defining the high predictive seismic parameters (HPSP) set. To validate the obtained results, the most effective model predictions have been compared with the results of the parametric step-by-step analyses performed for a new group of natural ground motions. The findings demonstrate that with a properly calibrated training phase, considering the specific site hazard and selecting seismic parameters from the HPSP set, the ML model can accurately estimate seismic responses whit a significantly reduced computational effort. This study underscores the potential of integrating ML techniques into the performance-based seismic design approach.

关键词： machine learning algorithm Artificial excitation Non-linear SDOF system Seismic parameter Peak relative displacement Predictive metrics

来源：评论

学校读者我要写书评

暂无评论

A Cost-effective and machine-learning-based method to identify and cluster redundant mutants in software mutation testing

引用

JOURNAL OF SUPERCOMPUTING 2024年第12期80卷 16711-16743页

作者： Arasteh, Bahman Ghaffari, Ali Istinye Univ Fac Engn & Nat Sci Dept Software Engn Istanbul Turkiye Appl Sci Private Univ Appl Sci Res Ctr Amman Jordan

The quality of software test data is assessed through mutation testing. This technique involves introducing various modifications (mutants) to the original code of the program. The test data's effectiveness, known as the test score, is quantified by the proportion of mutants that are successfully detected. A prominent issue within software mutation testing is the generation of an excessive quantity of mutants in programs. The primary objectives of this research are to diminish the total count of mutants by consolidating those that are duplicative, to decrease the overall mutation testing time by lessening the quantity of mutants produced, and to lower the expenses associated with mutation testing. This research introduces a machine learning-based strategy to recognise and eliminate redundant mutants. Building a machine learning-based classifier to classify the instructions according to the rate of error propagation is the first contribution of this study. Next, the remaining instructions of the source code are analyzed by the designed parser to generate single-line mutants. Unlike traditional approaches, mutants are not generated as distinct full-fledged programs. Instead, mutants consisting of a single line are selectively run using a developed instruction evaluator. Following this, a clustering technique is employed to categorise single-line mutants yielding identical outcomes into groups, where only one complete execution is needed per group. Testing on Java benchmarks with the new method has shown a decrease in the mutant count by 56.33% and a time reduction of 56.71% when compared with parallel tests using the MuJava and MuClipse tools. Despite the marked decrease in both mutant count and testing time, the mutation score remained consistent. Comparable outcomes were also observed with other mutation testing tools such as Pitest, Jester, Jumble, and JavaLancer.

关键词： Software mutation testing machine learning algorithm Single-line mutants Redundant mutant Cost of mutation test

来源：评论

学校读者我要写书评

暂无评论

Optimizing Wire Electrical Discharge Machining Parameters for Enhanced Gear Machining Performance: A machine learning Approach on 20MnCr5 Steel

引用

JOURNAL OF MATERIALS ENGINEERING AND PERFORMANCE 2024年 1-13页

作者： Vijayakumar, N. Chandradass, J. SRM Inst Sci & Technol Fac Engn & Technol Ctr Automot Mat Sch Mech EngnDept Automobile Engn Kattankulathur Campus Chennai 603203 Tamil Nadu India

Gears are machine elements that require high precision and surface quality to attain elevated performance of machinery. Many traditional techniques are capable of manufacturing gears but still need to overcome many drawbacks that require post-processing treatments to overcome. Wire electric discharge machining is an inevitable machining process for making gears with superior properties and quality characteristics. The current study explores the optimization of WEDM process parameters by hybridizing Taguchi-Multi criteria decision making-Supervised machine learning techniques toward output responses material removal rate, profile error, and surface roughness for 20MnCr5 steel material Five different WEDM parameters such as arc on time, arc off time, servo voltage, wire tension and wire feed rate have been varied in three levels to develop the experimental design matrix through Taguchi's orthogonal array with 27 experimental trails. The output responses have been analyzed through a signal-to-noise ratio method to identify optimal parameter settings and further analyzed through the CRITIC-TOPSIS technique for ranking the alternatives. The closeness coefficient values are analyzed through ANOVA and found that arc on time (71.38%) and wire tension (16.51%) have significant contributions toward combined objectives. Both decision tree and kNN algorithms have achieved a better classification accuracy of more than 90% indicating their potential in predicting future outcomes at 75:25 split ratio. Arc on time is the most significant factor in all the output responses studied as per response table rankings, ANOVA results, and decision tree structure. This study demonstrated the efficiency of combining experimental data with MCDM and ML to optimize WEDM parameters for enhanced gear machining.

关键词： 20MnCr5 steel machine learning algorithm MRR profile error surface roughness WEDM

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：