Purpose-The purpose of this paper is to introduce an evaluation methodology for employee profiles that will provide feedback to the training decision makers.Employee profiles play a crucial role in the evaluation proc...Purpose-The purpose of this paper is to introduce an evaluation methodology for employee profiles that will provide feedback to the training decision makers.Employee profiles play a crucial role in the evaluation process to improve the training process performance.This paper focuses on the clustering of the employees based on their profiles into specific categories that represent the employees’characteristics.The employees are classified into following categories:necessary training,required training,and no training.The work may answer the question of how to spend the budget of training for the employees.This investigation presents the use of fuzzy optimization and clustering hybrid model(data mining approaches)as a fuzzy imperialistic competitive algorithm(FICA)and k-means to find the employees’categories and predict their training requirements.Design/methodology/approach-Prior research that served as an impetus for this paper is discussed.The approach is to apply evolutionary algorithms and clustering hybrid model to improve the training decision system directions.Findings-This paper focuses on how to find a good model for the evaluation of employee profiles.The paper introduces the use of artificial intelligence methods(fuzzy optimization(FICA)and clustering techniques(K-means))in management.The suggestion and the recommendations were constructed based on the clustering results that represent the employee profiles and reflect their requirements during the training courses.Finally,the paper proved the ability of fuzzy optimization technique and clustering hybrid model in predicting the employee’s training requirements.Originality/value-This paper evaluates employee profiles based on new directions and expands the implication of clustering view in solving organizational challenges(in TCT for the first time).展开更多
Massive multiple-input multiple-output(MIMO)and intelligent reflecting surface(IRS)technologies have become a research focus for non-stationary vehicle-to-vehicle(V2V)wireless communications due to their capability to...Massive multiple-input multiple-output(MIMO)and intelligent reflecting surface(IRS)technologies have become a research focus for non-stationary vehicle-to-vehicle(V2V)wireless communications due to their capability to control radio propagation environment.In this paper,a non-stationary irregular geometry-based stochastic model(I-GBSM)for V2V massive MIMO systems using three-dimensional uniform linear arrays and discrete IRS at millimetre-wave operating frequencies is proposed.A new approach for determining IRS elements phase-shift using the Doppler effect and channel impulse response is introduced to mitigate channel non-stationarity and enhance propagation conditions.Unlike traditional models,it takes into account practical spherical wavefronts instead of plane wavefronts.The proposed model categorizes clusters into moving and static clusters to examine traffic density and its effects on channel characteristics in V2V environments.It employs a novel birth-death process to ensure consistency in cluster evolution.The non-stationary stochastic channel characteristics are comprehensively analyzed through simulations.These characteristics include space-time-frequency correlation functions,Doppler power spectral density,path loss,delay spread,root mean square error of the correlation function,and achievable rate across different operating frequencies.The analysis demonstrates notable performance improvements.The proposed I-GBSM is also validated by a good agreement with the results from existing models and measurements under reduced scenarios.展开更多
基金The authors thank the anonymous referees whose comments will help considerably to improve this paper.
文摘Purpose-The purpose of this paper is to introduce an evaluation methodology for employee profiles that will provide feedback to the training decision makers.Employee profiles play a crucial role in the evaluation process to improve the training process performance.This paper focuses on the clustering of the employees based on their profiles into specific categories that represent the employees’characteristics.The employees are classified into following categories:necessary training,required training,and no training.The work may answer the question of how to spend the budget of training for the employees.This investigation presents the use of fuzzy optimization and clustering hybrid model(data mining approaches)as a fuzzy imperialistic competitive algorithm(FICA)and k-means to find the employees’categories and predict their training requirements.Design/methodology/approach-Prior research that served as an impetus for this paper is discussed.The approach is to apply evolutionary algorithms and clustering hybrid model to improve the training decision system directions.Findings-This paper focuses on how to find a good model for the evaluation of employee profiles.The paper introduces the use of artificial intelligence methods(fuzzy optimization(FICA)and clustering techniques(K-means))in management.The suggestion and the recommendations were constructed based on the clustering results that represent the employee profiles and reflect their requirements during the training courses.Finally,the paper proved the ability of fuzzy optimization technique and clustering hybrid model in predicting the employee’s training requirements.Originality/value-This paper evaluates employee profiles based on new directions and expands the implication of clustering view in solving organizational challenges(in TCT for the first time).
基金supported by the National Natural Science Foundation of China(Grant No.W2433153)the National Natural Science Foundation of China under the State Major Research Instrument Program(Grant No.62027805)the Chaoyong Program(Grant No.130000-171207723/042),and the Startup Funds of Zhejiang University.
文摘Massive multiple-input multiple-output(MIMO)and intelligent reflecting surface(IRS)technologies have become a research focus for non-stationary vehicle-to-vehicle(V2V)wireless communications due to their capability to control radio propagation environment.In this paper,a non-stationary irregular geometry-based stochastic model(I-GBSM)for V2V massive MIMO systems using three-dimensional uniform linear arrays and discrete IRS at millimetre-wave operating frequencies is proposed.A new approach for determining IRS elements phase-shift using the Doppler effect and channel impulse response is introduced to mitigate channel non-stationarity and enhance propagation conditions.Unlike traditional models,it takes into account practical spherical wavefronts instead of plane wavefronts.The proposed model categorizes clusters into moving and static clusters to examine traffic density and its effects on channel characteristics in V2V environments.It employs a novel birth-death process to ensure consistency in cluster evolution.The non-stationary stochastic channel characteristics are comprehensively analyzed through simulations.These characteristics include space-time-frequency correlation functions,Doppler power spectral density,path loss,delay spread,root mean square error of the correlation function,and achievable rate across different operating frequencies.The analysis demonstrates notable performance improvements.The proposed I-GBSM is also validated by a good agreement with the results from existing models and measurements under reduced scenarios.
