Integrating non-targeted metabolomics and machine learning for comprehensive phytochemical profiling and intelligent discrimination of Acorus tatarinowii and its adulterants

Key Points

• Non-targeted metabolomics successfully distinguishes Acorus tatarinowii from its adulterants, enhancing identification.
• The model shows an impressive accuracy rate above 90%, confirming reliability in phytochemical profiling.
• This analysis utilizes machine learning to improve the identification of complex phytochemical compositions in samples.
• Such methods may enable better protection against fraudulent practices in herbal product markets.