A Hybrid Electronic Components Detection Parameter Sequence Optimization Method based on Random Forest and Mean Decrease Accuracy

Aiming at the problems in electronic components detection parameter setting, this paper proposes an optimization method of based on Random Forest and Mean Decrease Accuracy. Firstly, use data preprocessing method to deal with the collected data and make these data suitalbe for random forest model, then use random forest model based on mean decrease accuracy to optimize detection parameter sequence, find optimal detection parameter sequencing scheme, shorten the detection time and reduce the cost, and finally improve the detection accuracy, effectiveness and economy. The cases analysis shows that the proposed method in this paper has a higher digital recognition rate, fewer detection times and shorter time in the same batch of products, indicating the effectiveness and applicability of this method.