A working pattern recognition model of airborne fire control radar for unbalanced data is proposed in this paper, which consists of weight oversampling and proportional voting random forest. Generally, there are two m...
详细信息
A working pattern recognition model of airborne fire control radar for unbalanced data is proposed in this paper, which consists of weight oversampling and proportional voting random forest. Generally, there are two methods for solving the problem of the unbalanced data: data preprocessing and integrated learning. Most of the current methods have the problems such as data redundancy during oversampling, overfitting during identification, and unreasonable decision-making principles. The proposed model is made of the weight oversampling algorithm in data processing, and the proportional voting random forest in recognition algorithm. Simulation results show that, compared with traditional methods, the model proposed in this paper has greatly improved the recognition effect and recognition efficiency. It solves the problem of radar working pattern recognition in the case of unbalanced samples to a certain extent.
暂无评论