To address the issue that single traffic data missing value imputation models cannot comprehensively handle the multi-source heterogeneity and complex data volume of traffic data,a multi-classifier imputation model based on adaptive weighting determined by information entropy was proposed. First,information entropy representing "disorder degree" was introduced to evaluate prediction quality and determine multi-classifier weights. Second,a dynamic adaptive weighting method was designed to resolve the problem of different classifiers being suitable for various samples caused by device heterogeneity. Finally,validation was conducted on both public and self-collected datasets. The results demonstrate that the proposed model achieves significant improvement in detection performance compared with other imputation models. It also attains high accuracy in experiments on the public Interstate Highway Traffic Flow Dataset,with an F1 of 0778 and a 10% improvement in RMSE, exhibiting strong generalizability. By enabling weights to adaptively evolve with data streams based on information entropy determination,the algorithm achieves faster detection speed and higher accuracy,providing technical references for the establishment of missing value imputation models in traffic data cleaning.