The objective characterization of food odor is important for the optimization of food production process and quality evaluation. However， the complexity of the food odor formation mechanism and its complicated composition， coupled with the fact that the odor evaluation process is susceptible to a variety of factors such as environmental， psychological and perception modes， make it necessary to deal with a large amount of high-dimensional and complex information in the characterization of food odors， which requires the use of tools with powerful data processing capabilities. Along with the rapid development of artificial intelligence， computer science and big data technology， machine learning has powerful data processing capability， it has become possible to characterize food odors using machine learning， and the implementation of this process relies heavily on machine learning algorithms. In this paper， the application of various machine learning methods in food odor characterization was summarized， and the advantages and problems to be solved in using machine learning methods for food odor characterization were pointed out. Finally， the future development direction of machine learning algorithms applied to food odor characterization was foreseen.