A Random Forest Algorithm-Based Model for Assessing Brands’ Digital Marketing Capabilities
Yue Zhang · 2024
This paper constructs a brand digital marketing capability assessment model based on the random forest algorithm. First, relevant marketing data indicators are obtained by constructing an input feature set covering multiple dimensions such as brand influence and market coverage. Then, the random forest algorithm is utilized to assess these indicators and derive the corresponding scores. By means of weighted average, the assessment results of different indicators are synthesized, and the comprehensive evaluation of the brand's digital marketing capability is finally obtained. In the process of model construction, the weighted scoring method is used, combined with the fuzzy priority relationship matrix and the affiliation function number and other methods, to effectively determine the weight of each assessment index. By comparing the performance of the random forest algorithm and the neural network algorithm in the assessment process, the results show that the random forest algorithm performs superiorly in the number of scenarios in multiple output intervals, showing its efficiency and accuracy in the assessment of the brand's digital marketing capability. Finally, the effectiveness of the model is further verified by analyzing the weights of the assessment index system, which provides a scientific basis for the digital transformation of the brand.