Search

Objective A novel technique for the identification of gas-to-liquid (GTL) lubricants based on mutual information method and Bayesian algorithm is proposed to enhance the tariff identification of lubricating oils. Methods First, feature parameters of GTL lubricants and conventional petroleum lubricants are extracted, and the mutual information algorithm is used to select feature indicators with discriminative ability. Then, the Bayesian algorithm is used to model the selected feature indicators, which finally achieves the classification discrimination of GTL lubricants and conventional lubricants. Results The experimental results demonstrate the effectiveness of the proposed method in accurately discriminating between GTL lubricants and conventional petroleum lubricants. Conclusions The established model exhibits good stability and reliability, providing a comprehensive and accurate identification tool for the lubricating oil industry and tariff assessment. [ABSTRACT FROM AUTHOR]

Aiming at the shortcomings of normal SPBO, this paper analyzed the psychological characteristics of students ' learning, and put forward an improved student psychology based optimization algorithm ( HSSPBO) using hybrid strategy . Firstly, it took the inverse value of student ' s test scores as the fitness value of the student, and took the fitness value of the best students in the class as the benchmark, and then divided the students in the class into four categories : the best students, the good students, the ordinary students and the students who tried random improvement. Secondly, it used the dynamic switching probability of sine square and cosine square to balance global exploration and local development, so that the global exploration ability and local exploitation ability of the algorithm were improved effectively. Thirdly, by introducing Cauchy mutation strategy to change the local search step size, it improved effectively the local search ability of the algorithm and enhanced the ability of the algorithm to jump out of the local optimization. Finally, the HSSPBO algorithm used Levy flight strategy to make the individual search step length more random and flexible, which effectively enhanced the individual ' s searching ability, and further improved the optimization speed of the algorithm. Through simulation experiments of 12 benchmark functions and comparison with six optimization algorithms, the results show that HSSPBO ' s global search ability has been significantly improved, and it has faster global convergence speed, better optimization accuracy and stability in function optimization. [ABSTRACT FROM AUTHOR]

In recent years, as deep learning has achieved good results in many fields, deep learning has also begun to be applied to recommendation systems, such as NFM models and DeepFM models that use deep learning technology to capture high-level feature interactions. However, considering the changes in the external environment and internal perception, the user' s interest should also change dynamically over time, and the combination based on the original features may not necessarily learn effective feature interaction. This paper attempted to build a new model FG_DRFwFm, which could learn the interaction of low-level and high-level features of multiple feature domains and dealt with long-term changes in user interest. It constructed the training features by constructing new features based on the original features and splicing them together, which could better learn effective feature interaction. Finally, the proposed model compared the recommendation effect with multiple advanced CTR algorithms on the MovieLens data set. The experimental results show that the proposed model achieves better results. [ABSTRACT FROM AUTHOR]

The Aquila optimizer（AO), although capable of robust global exploration, had problems with inadequate local development. In the study, a differential evolution mutation and tangent flight Aquila optimizer（DEtanAO） was proposed. Firstly, the mutation operation in the differential evolution algorithm can make the algorithm have strong development ability and make up for the shortcomings of the AO algorithm. Then, the tangent flight strategy in the tangent search algorithm has a strong ability to explore the search space and can make the algorithm jump out of the local optimal solution, which is used to replace the Levy flight in the AO algorithm. The combination of these two strategies effectively balanced the exploration and exploitation stage of the DEtanAO algorithm. Finally, in order to verify the optimization performance of the DEtanAO algorithm, the optimization ability of the improved algorithm was tested in 12 standard benchmark functions, highdimensional functions, Wilcoxon rank-sum test, and engineering optimization problems. The results showed that the optimization ability and convergence speed of the DEtanAO algorithm was better, compared with other newly proposed intelligent algorithms. [ABSTRACT FROM AUTHOR]