基于随机森林的猪肉价格预测模型

Pork Price Prediction Model Based on Random Forest

摘要: 中国是世界上最大的猪肉生产国和消费国，也是生猪养殖规模最大的国家。在中国的所有肉类消费中，猪肉消费一直领先其他肉制品，近年来生猪价格波动频繁，会对整个养猪业乃至社会造成巨大影响。因此探究一种可以准确预测猪肉价格的模型对生猪市场的研究和生产都具有重要意义。随机森林是以K个决策树为基本分类器，进行集成学习后得到的组合分类器，可以解决数据多模态问题。考虑到猪肉价格与其他因素之间的复杂多模态非线性关系，故本文使用随机森林对猪肉价格进行预测。针对收集的猪肉价格影响因素(如玉米价格，牛肉价格等)，建立多棵决策树构建随机森林模型，对猪肉价格实现精准预测。同时进行了对比实验，对比决策树、支持向量机预测模型，实验结果表明基于随机森林的预测价格数据和真实价格数据拟合效果最好。

Abstract: China is the world’s largest producer and consumer of pork, and it is also the country with the largest scale of pig farming. Among all meat consumption in China, pork consumption has always been ahead of other meat products. In recent years, the price of live pigs has fluctuated frequently, which will have a huge impact on the entire pig industry and society. Therefore, exploring a model that can accurately predict the price of pork is of great significance to the research and production of the live pig market. Random forest is based on K decision trees as the basic classifier, and the combined classifier is obtained after ensemble learning, which can solve the data multi-modal problem. Considering the complex multi-modal nonlinear relationship between pork price and other factors, this paper uses random forest to predict pork price. According to the collected pork price influencing factors (such as corn price, beef price, etc.), establish multiple decision trees to construct a random forest model to realize accurate prediction of pork price. At the same time, a comparative experiment was carried out to compare the prediction model of decision tree and support vector machine. The experimental results showed that the predicted price data based on random forest and the real price data have the best fit.