Smell threshold prediction method and device and household appliance

The invention belongs to the field of data processing, and discloses a smell threshold prediction method, which comprises the following steps: acquiring chemical feature information of a peculiar smell substance to be predicted; inputting the chemical feature information into a trained smell threshold value prediction model for processing, outputting a smell threshold value corresponding to the to-be-predicted odor substance, wherein the smell threshold value prediction model is obtained through training according to sample data of which the chemical feature information and the smell threshold value are determined. By inputting the chemical characteristic information of the peculiar smell substance into the trained smell threshold prediction model, the acquisition efficiency of the smell threshold is improved, the influence of test conditions and individual differences is avoided, and the precision of the smell threshold is improved. 本申请属于数据处理领域，公开了一种嗅味阈值预测方法，包括：获取待预测的异味物质的化学特征信息；将所述化学特征信息输入至已训练的嗅味阈值预测模型中处理，输