Aiming at the model overfitting problem caused by limited datasets and visual complexity in succulent plant classification tasks, this study proposes a GoogLeNet classification method based on lightweighting and improving the Convolutional Block attention module (CBAM). Meanwhile, batch normalization (BN) operations are added after each convolutional layer to accelerate network convergence and improve model stability. In addition, the model’s ability to extract key features is enhanced by integrating the channel and spatial attention mechanisms of the CBAM attention module. Experimental results show that the improved lightweight GoogLeNet achieves 99.4% classification accuracy on the validation set, effectively mitigates the overfitting problem, and maintains high computational efficiency in resource-constrained environments. The model parameters and computational complexity are significantly reduced by streamlining the Inception modules from nine to seven and introducing depth-separable convolution. To further validate the model robustness, this study extends the dataset via data augmentation methods, and the experiments show that the improved model still maintains stable performance in small dataset environments, demonstrating its advantages in data-scarce scenarios. This study provides an effective solution for the task of succulent plant classification, which has significant application value. Future research will focus on further optimization of the model structure to continuously improve the classification accuracy and robustness.

Aiming at the problems of massive heterogeneous data in smart university, this study designed and developed a big data platform for university based on Hadoop. We firstly used HBase and Hive SQL to store and analyze the collected datasets. Subsequently, the corresponding results were visualized using ECharts. This big data center with powerful capabilities of integration, storage and analysis can realize the most basic and core data "management + use". Furthermore, it would provide scientific decision-making services for the daily management of university and promotes the rapid development of campus informatization.

Metamorphosis is one of the major innovations in the evolutionary history of insect, playing an important role in environmental adaptation and biodiversity formation. However, it is still a challenge to recognize the underlying mechanisms of insect metamorphosis in traditional molecular biology. In this study, based on 138 insect genomes across 10 orders, we present a gene network classification model (GNCM) utilizes deep learning (DL) and attention mechanism (AM) to mine some key genes of metamorphosis. The GNCM identified a total of 97 sets of orthologous genes (OGs) with high weights in insect metamorphosis, and the subsequent annotations suggested that these genes were involved in hormone regulation, immune, detoxification, metabolism and developmental functions. Notably, 38 genes with high weights showed specific expression lever of their transcripts during the pupal stage. This study not only developed a new method of artificial intelligence for evolutionary biology, but also provides a new insight to understand the macroevolutionary of insect metamorphosis.