Seabuckthorn seed oil has high nutritional and medicinal value, and high profits drive illegal businesses to pass shoddy goods off as good, confusing the false with the true. More methodological research is needed to identify seabuckthorn seed oil adulteration. Therefore, this paper proposes a method to detect the adulteration of seabuckthorn seed oil based on the FA-XGBoost algorithm, which enables Raman spectroscopy. Seabuckthorn seed oil was adulterated with sunflower seed oil in different volume fractions. The Raman spectra of all samples were measured by a near-infrared micro Raman spectrometer. The Raman spectra of seabuckthorn seed oil and sunflower seed oil were qualitatively analyzed. A regression model for adulterating sunflower seed oil with seabuckthorn seed oil was constructed using the FA-XGBoost algorithm. The predictive performance of the FA-XGBoost model was evaluated using a test set with a determination coefficient of 0.9959 and a mean squared error of 0.0031. This paper proposes a quantitative detection method for seabuckthorn seed oil adulteration. The concentration prediction of sunflower seed oil adulterated with seabuckthorn seed oil is realized through the combination of the FA-XGBoost algorithm and Raman spectroscopy. This method has potential application value and practical significance for regulating the domestic seabuckthorn seed oil consumption market.