Abstract: 【Objective】 The objective of this study was to evaluate the quality of traditional yak ghee and provide reference for food industry inspection and product quality optimization.【Method】 The acid value,peroxide value,and odor information of yak ghee samples were analyzed using the Least Significant Difference（LSD） test. The place of production and origin of yak ghee were qualitatively and quantitatively analyzed using Principal Component Analysis（PCA）,Canonical Discriminant Analysis（CDA） and MultiLayer Perceptron（MLP）.【Result】 The acid value（4. 35～4. 71 mg/g） was found to be higher than the national standard,while the peroxide value（0. 000～0. 0075 g/100g） fell within the required range. The quality of yak ghee varied among different families. PCA analysis successfully distinguished the sources of yak ghee samples from 16 families. In CDA analysis,the data points of each sample were more concentrated,resulting in larger distances between sample groups and a stronger discrimination effect compared to PCA.However,there were still some overlapping data points. MLP was employed for quantitative identification of the origin of traditional handmade yak ghee and traceability. The MLP 10-1-2 grid structure showed that the average accuracy,accuracy,sensitivity,specificity,and area under the curve for yak ghee in Luqu and Maqu were 99. 51%,99. 53%,99. 53%,99. 50%,and 99. 52%,respectively. The MLP 10-1-16 grid structure showed that the average accuracy,accuracy,sensitivity,specificity,and area under the curve for yak ghee from 16 herdsmen were 98. 97%,92. 18%,92. 23%,99. 45%,and 95. 84%,respectively.【Conclusion】 Based on the findings of this study,the quality of traditional yak ghee can be effectively evaluated using E-nose technology.