This study utilizes NASA’s Normalized Difference Vegetation Index (NDVI) data from the Google Earth Engine (GEE) platform and employs methods such as mean analysis, trend analysis, and the Hurst index to assess NDVI dynamics in Xinjiang, with a particular focus on desert, meadow, and grassland vegetation. Furthermore, multiple linear regression, random forest, support vector machines, and XGBoost models are applied to construct and evaluate the NDVI prediction models. The key driving forces are identified and ranked based on the results of the optimal model. Changes in the vegetation cover in response to these driving forces are analyzed using the Mann–Kendall test and partial correlation analysis. The results indicate the following: (1) From 2000 to 2023, the annual variation in NDVI in Xinjian fluctuates at a rate of 0.0012 per year. The intra-annual trend follows an inverted U shape, with meadow vegetation exhibiting the highest monthly NDVI fluctuations. (2) During this period, the annual average NDVI in Xinjiang ranges from 0 to 0.3, covering 74.74% of the region. Spatially, higher NDVI values are observed in the north and northwest, while lower values are concentrated in the south and southeast. (3) The overall slope of the variation in NDVI in Xinjiang between 2000 and 2023 ranges between −0.034 and 0.047, indicating no significant upward trend. According to the Hurst index, future projections suggest a shift from vegetation improvement to potential degradation. (4) Machine learning models are developed to predict NDVI, with random forest and XGBoost showing the highest precision. Soil moisture, runoff, and potential evaporation are identified as key drivers. In the last 24 years, the temperatures in Xinjiang have generally increased, while precipitation, soil moisture, and runoff have declined. There is a significant negative correlation between NDVI and both temperature and potential evaporation, while the correlation between NDVI and precipitation, soil moisture, and runoff is positive and significant, with distinct spatial variations throughout the region. The overall trend of vegetation cover in Xinjiang has been increasing, but the future outlook is less promising. Enhanced environmental monitoring and protective measures are essential moving forward.

Livestock health knowledge graphs are essential for decision-making and reasoning in animal husbandry, yet existing knowledge is scattered across unstructured literature and encoded in narrowly scoped, species-specific models, resulting in semantic fragmentation and limited reusability. To address these issues, we proposed FARM (Four-dimensional Automated-Reasoning Multi-agent), a zero-shot multi-agent framework used for constructing multi-species livestock health knowledge graphs. FARM is grounded in a Four-Dimension Livestock Health Framework encompassing Rearing Environment, Physiological Status, Feed & Water Inputs, and Production Performance, and employs a unified ontology strategy that integrates cross-species general labels with species-specific constraints to achieve semantic alignment. The framework orchestrates five specialized agents—Coordination, Entity Extraction, Ontology Normalization, Relation Extraction, and Knowledge Fusion—to automate the construction process. Experiments on 2478 expertly annotated text samples demonstrate that FARM achieves an entity-level F1 score of 0.8070 (IoU ≥ 0.5), surpassing the strongest baseline by 0.1627. Moreover, it attains a corrected entity label accuracy of 90.44% and an F1 score of 0.9277 in relation existence identification, outperforming the baseline by 0.1114. Validation on 500 image samples further confirms its capability in multimodal evidence fusion. The resulting knowledge graph contains 29,064 entities and 26,662 triples, providing a reusable foundation for zero-shot extraction and unified cross-species semantic modeling.

With the rapid advancement of artificial intelligence, intelligent farming has become a key trend in modern agriculture. In particular, the application of intelligent systems in broiler farming is essential for enhancing production efficiency and optimizing management practices. Broiler farming is a complex process involving multiple interrelated components. However, existing knowledge graphs primarily focus on disease and prevention, making it difficult to capture the intricate interdependencies within the farming process. This limits the effectiveness of knowledge-based support in decision-making. To develop a high-quality broiler farming knowledge system, this study adopts large language modeling technology to integrate a Chinese corpus and construct a comprehensive knowledge graph dataset covering four core dimensions: broiler breeds, farming environment, feeding management, and disease prevention.