Plant modeling uses mathematical and computational methods to simulate plant structures, physiological processes, and interactions with various environments. In precision agriculture, it enables the digital monitoring and prediction of crop growth, supporting better management and efficient resource use. Wheat, as a major global staple, is vital for food security. However, wheat stripe rust, a widespread and destructive disease, threatens yield stability. The paper proposes wheat cultivation suitability evaluation with stripe rust disease using an agriculture group consensus framework (WCSE-AGC) to tackle this issue. Assessing stripe rust severity in regions relies on wheat pathologists' judgments based on multiple criteria, creating a multi-attribute, multi-decision-maker consensus problem. Limited regional coverage and inconsistent evaluations among wheat pathologists complicate consensus-reaching. To support wheat pathologist participation, this study employs artificial-intelligence-generated content (AIGC) techniques by using Claude 3.7 to simulate wheat pathologists' scoring through role-playing and chain-of-thought prompting. WCSE-AGC comprises three main stages. First, a graph neural network (GNN) models trust propagation within wheat pathologists' social networks, completing missing trust links and providing a solid foundation for weighting and clustering. This ensures reliable expert influence estimations. Second, integrating secretary bird optimization (SBO), K-means, and three-way clustering detects overlapping wheat pathologist subgroups, reducing opinion divergence and improving consensus inclusiveness and convergence. Third, a two-stage optimization balances group fairness and adjustment cost, enhancing consensus practicality and acceptance. The paper conducts experiments using publicly available real wheat stripe rust datasets from four different locations, Ethiopia, India, Turkey, and China, and validates the effectiveness and robustness of the framework through comparative and sensitivity analyses.
Keywords: artificial intelligent generated content; plant disease detection; precision agriculture; wheat cultivation; wheat strip rust disease.