Multimodel fusion and decision optimization framework for high-dimensional detection data
Abstract
To address core challenges in high-dimensional complex detection data—such as characterizing nonlinear relationships, enabling multi-factor collaborative decision-making, and achieving precise grouping—this paper constructs a comprehensive analytical framework based on big data and algorithmic technologies. First, a hybrid nonlinear statistical model is established to quantify the intricate associations between target variables and key influencing factors, with model validity verified through statistical testing. Second, the K-means clustering algorithm is employed for precise sample grouping, determining optimal decision points based on intra-group homogeneity. Third, a multi-objective statistical optimization model is constructed to comprehensively quantify multiple risk categories and achieve balanced target optimization. Finally, a logistic regression classification model completes state determination, outputting classification results with probability support. Experiments conducted on large-scale real-world datasets demonstrate that the proposed framework exhibits outstanding performance in variable interpretability, grouping rationality, decision optimization capability, and classification accuracy. It provides a systematic and actionable algorithmic solution for data analysis and decision-making in similar scenarios.