Big data-driven intelligent analysis model of enterprise strategic planning and market competition

In the context of digital transformation, enterprises urgently need strategic planning and competition analysis tools based on big data. This paper aims to build an intelligent analysis model integrating random forest algorithms to improve market competition prediction accuracy and decision-making reliability. This method deals with multi-source heterogeneous data, applies information gain and Gini coefficient to optimize feature selection, and adopts bagging voting integration to realize high-dimensional data analysis. The empirical results show that the model achieves an accuracy rate of 93.2% on the test set and maintains a robustness of 91.7% in a 30% noise environment, increases the adoption rate of enterprise strategic recommendations to 82.6% and enables the enterprise to achieve quarterly revenue growth of 4.2 percentage points Meanwhile, the feature interpretability analysis of the model is highly consistent with expert assessment (ρ = 0.92). The above conclusions prove that the model significantly optimizes the dynamic decision-making efficiency of enterprises, but it is necessary to further strengthen the unstructured data processing ability through multi-modal fusion and incremental learning.