In agriculture, crop yield and quality are critical for global food supply and human survival. Challenges such as plant leaf diseases necessitate a fast, automatic, economical, and accurate method. This paper utilizes deep learning, transfer learning, and specific feature learning modules (CBAM, Inception-ResNet) for their outstanding performance in image processing and classification. The ResNet model, pretrained on ImageNet, serves as the cornerstone, with introduced feature learning modules in our IRCResNet model. Experimental results show our model achieves an average prediction accuracy of 96.8574% on public datasets, thoroughly validating our approach and significantly enhancing plant leaf disease identification.

This study aims to examine the pathways through which the user experience (UX) of ChatGPT, a representative of generative artificial intelligence, affects user loyalty. Additionally, it seeks to verify whether ChatGPT’s UX varies according to a user’s need for cognition (NFC). This research proposed and examined how ChatGPT’ UX affect user engagement and loyalty and used mediation analysis using PROCESS Macro Model 6 to test the impact of UX on web-based ChatGPT loyalty. Data were collected by an online marketing research company. 200 respondents were selected from a panel of individuals who had used ChatGPT within the previous month. Prior to the survey, the study objective was explained to the respondents, who were instructed to answer questions based on their experiences with ChatGPT during the previous month. The usefulness of ChatGPT was found to have a significant impact on interactivity, engagement, and intention to reuse. Second, it was revealed that evaluations of ChatGPT may vary according to users’ cognitive needs. Users with a high NFC, who seek to solve complex problems and pursue new experiences, perceived ChatGPT’s usefulness, interactivity, engagement, and reuse intentions more positively than those with a lower NFC. These results have several academic implications. First, this study validated the role of the UX in ChatGPT. Second, it validated the role of users’ need for cognition levels in their experience with ChatGPT.

This paper uses quantitative research methods to explore the differences in the impact of virtual influencers on different consumer groups in the context of technological integration and innovation. The study uses DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering technology to segment consumers and combines social media behavior analysis with purchase records to collect data to identify differences in consumer behavior under the influence of virtual influencers. Consumers’ emotional resonance and brand awareness information about virtual influencers are extracted through sentiment analysis technology. The study finds that there are significant differences in the influence of virtual influencers on different consumer groups, especially in high-potential purchase groups, where the influence of virtual influencers is strong but short-lived. This paper further explores the deep integration of virtual influencer technology with new generation information technologies such as 5G and artificial intelligence, and emphasizes the importance of such technological integration in enhancing the endogenous and empowering capabilities of virtual influencers. The research results show that technological integration and innovation can not only promote the development of virtual influencers, but also provide new technical support for infrastructure construction, especially in the fields of smart cities and industrial production. This paper provides a new theoretical perspective for the market application of virtual influencers and provides practical support for the application of virtual technology in infrastructure construction.