摘要：【中文】本发明涉及分类模型领域，公开了一种数据预处理方法、装置、计算机设备及存储介质，其方法包括：判断天气样本数据的指定因子的数据类型；获取与数据类型匹配的第一数据填充方法，并根据第一数据填充方法对天气样本数据在指定因子上的缺失值进行第一次填充；在所有天气样本数据的缺失值填充完毕后，通过预设嵌入式特征选择算法对存在已填充缺失值的指定因子进行筛选，获得优选因子；判断已填充的缺失值属于优选因子的天气样本数据是否存在历史数据；若已填充的缺失值属于优选因子的天气样本数据存在历史数据，则根据第二数据填充方法和历史数据计算替换值，将已填充的缺失值替换为替换值。本发明可提高数据挖掘方法的预测准确率。 【EN】The invention relates to the field of classification models, and discloses a data preprocessing method, a data preprocessing device, computer equipment and a storage medium, wherein the method comprises the following steps: judging the data type of the designated factor of the weather sample data; acquiring a first data filling method matched with the data type, and filling missing values of the weather sample data on the designated factors for the first time according to the first data filling method; after the missing values of all weather sample data are filled, screening specified factors with filled missing values through a preset embedded feature selection algorithm to obtain preferred factors; judging whether historical data exist in the weather sample data with the filled missing values belonging to the preferred factors; and if the filled missing value belongs to the historical data of the weather sample data of the preferred factor, calculating a replacement value according to the second data filling method and the historical data, and replacing the filled missing value with the replacement value. The invention can improve the prediction accuracy of the data mining method.