Quality inspection is essential in preventing defective products from entering the market. Due to the typically low percentage of defective products, it is generally challenging to detect them using algorithms that aim for the overall classification accuracy. To help solve this problem, we propose an ensemble learning classification model, where we employ adaptive boosting (AdaBoost) to cascade multiple backpropagation (BP) neural networks. Furthermore, cost-sensitive (CS) learning is introduced to adjust the loss function of the basic classifier of the BP neural network. For clarity, this model is called a CS-AdaBoost-BP model. To empirically verify its effectiveness, we use data from home appliance production lines from Bosch. We carry out tenfold cross-validation to evaluate and compare the performance between the CS-AdaBoost-BP model and three existing models: BP neural network, BP neural network based on sampling, and AdaBoost-BP. The results show that our proposed model not only performs better than the other models but also significantly improves the ability to identify defective products. Furthermore, based on the mean value of the Youden index, our proposed model has the highest stability.
LaSrBO4(B=Cu、Ni、 Co、 Fe、 Mn、 Cr ) was studied with respect to their preparation, characteristics and catalytic oxidation activity by means of XRD, XPS, electric conductivity and activity evaluation. Results showed that the calcination temperature has great influence on the formation of the phases of such samples, and was associated with the radius of B ion. The greater the B ion radius, the higher the calcination temperature required for the formation of the phases. It was also found that the activity and the relative content of adsorptive oxygen and lattice oxygen on the sample surf are, the activation energy of electron conduction and the CO oxidation activity may vary with different B ions, and demonstrate a dual-peak pattern with the increase in the number of 3d electrons of the B ion. 相似文献
The Ramanujan Journal - Let $$k\ge 2$$ be an even integer, and let M be a positive integer. We prove a hybrid subconvexity bound for $$L(1/2,\text {sym}^2 f\otimes \chi )$$ with the associated... 相似文献
In this paper, we are interested in the regularity estimates of the nonnegative viscosity super solution of the $β$−biased infinity Laplacian equation $$∆^β_∞u = 0,$$ where $β ∈ \mathbb{R}$ is a fixed constant and $∆^β_∞u := ∆^N_∞u + β|Du|,$ which arises from the random game named biased tug-of-war. By studying directly the $β$−biased infinity Laplacian equation, we construct the appropriate exponential cones as barrier functions to establish a key estimate. Based on this estimate, we obtain the Harnack inequality, Hopf boundary point lemma, Lipschitz estimate and the Liouville property etc. 相似文献
4-Fluorinated UDP-MurNAc pentapeptide, 2, has been synthesized. In our previous study, UDP-MurNAc pentapeptide analogue 1 was found to be incorporated into the bacterial cell wall through biosynthesis. Compound 2 showed growth-inhibition activity against Gram-positive bacteria when it was added to growth media at 0.01 mg/mL. [structure--see text] 相似文献