The configuration of the hyperparameters in convolutional neural networks (CNN) is crucial for determining their performance. However, traditional methods for hyperparameter configuration, such as grid searches and random searches, are time consuming and labor intensive. The optimization of CNN hyperparameters is a complex problem involving multiple local optima that poses a challenge for traditional particle swarm optimization (PSO) algorithms, which are prone to getting stuck in the local optima and achieving suboptimal results. To address the above issues, we proposed an adaptive dimensional Gaussian mutation PSO (ADGMPSO) to efficiently select the optimal hyperparameter configurations. The ADGMPSO algorithm utilized a cat chaos initialization strategy to generate an initial population with a more uniform distribution. It combined the sine-based inertia weights and an asynchronous change learning factor strategy to balance the global exploration and local exploitation capabilities. Finally, an elite particle adaptive dimensional Gaussian mutation strategy was proposed to improve the population diversity and convergence accuracy at the different stages of evolution. The performance of the proposed algorithm was compared to five other evolutionary algorithms, including PSO, BOA, WOA, SSA, and GWO, on ten benchmark test functions, and the results demonstrated the superiority of the proposed algorithm in terms of the optimal value, mean value, and standard deviation. The ADGMPSO algorithm was then applied to the hyperparameter optimization for the LeNet-5 and ResNet-18 network models. The results on the MNIST and CIFAR10 datasets showed that the proposed algorithm achieved a higher accuracy and generalization ability than the other optimization algorithms, such as PSO-CNN, LDWPSO-CNN, and GA-CNN.

As the primary means of modern enterprise management, business process management (BPM) technology has become the mainstream development trend of modern enterprise management. The efficient and accurate establishment of business processes is essential for effective BPM. However, the traditional manual-based modeling approach is time-consuming and error-prone. To overcome this, process recommendation technology can improve the intelligence and efficiency of modeling to a certain extent. However, existing process modeling recommendation methods suffer from the problem of low accuracy and neglecting short-process models. Therefore, a novel process modeling recommendation method that integrates disjoint paths and sequential patterns was proposed. This method uses edge-disjoint paths for the first time to represent the behavioral semantics of processes, and an improved contiguous sequential pattern mining algorithm was proposed to mine the contiguous path sequential patterns (CPSPs) of edge-disjoint paths. In the process modeling recommendation stage, the k CPSPs with the highest matching degree with the current reference model process were calculated, and the last node in these CPSPs was used as the set of recommendation nodes. In cases with CPSPs with the same matching degree, the one with the higher value was recommended according to their corresponding lift, confidence, and support degrees. Through experimental evaluation and comparison, it was shown that the proposed method effectively improved the accuracy of the recommendation of both short-process and long-process models while ensuring effectiveness and time efficiency.