Research on Multidimensional Data Analysis and Decision Making Based on Multiple Machine Learning Algorithms

Focusing on multidimensional data analysis, this study applies a variety of machine learning algorithms for an in-depth exploration of decision preferences and influencing factors. A comprehensive empirical analysis is conducted through the entropy-weighted TOPSIS comprehensive evaluation method, K-Means clustering model, decision tree and random forest, demonstrating the superiority of the algorithms in complex data processing. First, the entropy weight-based TOPSIS evaluation method identifies and quantifies the key influencing factors and analyzes their weights. On this basis, the variance decomposition methods of Dagum's Gini coefficient decomposition model and VAR model are used to reveal the structural differences in the dataset. Next, the decision preference patterns in the data were analyzed using K-Means clustering and decision tree models. Finally, the random forest model further reveals the important influences behind the data decisions. This research provides a strong theoretical basis and practical tools for complex data analysis and decision support through the application of multiple machine learning algorithms.