基于随机森林的“拍照赚钱”定价方案分析

Discussion on the Pricing Scheme of “Making Money from Photography” Based on Random Forest

论文以2017年全国大学生数学建模竞赛B题“拍照赚钱”定价方案数据为研究对象,通过随机森林算法建立APP拍照任务数据的相关指标与拍照酬金之间的定量关系的数学模型,优化“拍照赚钱”定价方案。实验结果显示,用随机森林回归对任务价格进行重新标定后,新的任务价格范围是65.021~83.104元,平均价格为69.058元,标准差为3.955元;新的任务价格范围是原价格范围(65~85元)的0.904倍,比原平均价格(原平均价格69.111元)降低了0.053元,比原定价的标准差(原定价标准差4.513元)降低了0.558元,说明随机森林回归法作为“拍照赚钱”定价方案数学模型效果较好。该项研究为APP自助服务平台的新任务价格标定提供参考和借鉴。

This paper takes the pricing scheme of “making money from photography” in the B problems of 2017 Undergraduate Mathematical Modeling Contest of China as the research content, and establishes a mathematical model of the quantitative relationship between the relevant indicators of the APP Photo Task Data and the Photo Remuneration through the Random Forest Algorithms, and optimizes the pricing scheme of “Making money from photography”. The experimental results show that the new task price range is 65.021-83.104 yuan, the average price is 69.058 yuan, and the standard deviation is 3.955 yuan. The new task price range is 0.904 times of the original price range (65-85 yuan). The average price of the new task is 0.053 yuan lower than the original average price (69.111 yuan). The standard deviation is 0.558 yuan lower than the standard deviation of the original pricing (4.513 yuan). The results show that the Random Forest regression method is better as a mathematical model of “making money from photography” pricing scheme. This study provides a reference for the calibration of new task price of APP self-service platform.