Identification of candidate medicinal herbs for skincare via data mining of the classic Donguibogam text on Korean medicine

BACKGROUND: Korean cosmetics are widely exported throughout Asia. Cosmetics exploiting traditional Korean medicine lead this trend; thus, the traditional medicinal literature has been invaluable in terms of cosmetic development. We sought candidate medicinal herbs for skincare. METHODS: We used data mining to investigate associations between medicinal herbs and skin-related keywords (SRKs) in a classical text. We selected 26 SRKs used in the Donguibogam text; these referred to 626 medicinal herbs. Using a term frequency-inverse document frequency approach, we extracted data on herbal characteristics by assessing the co-occurrence frequencies of 52 medicinal herbs and the 26 SRKs. RESULTS: We extracted the characteristics of the 52 herbs, each of which exhibited a distinct skin-related action profile. For example Ginseng Radix was associated at a high-level with tonification and anti-aging, but Rehmanniae Radix exhibited a stronger association with anti-aging. Of the 52 herbs, 46 had been subjected to at least one modern study on skincare-related efficacy. CONCLUSIONS: We made a comprehensive list of candidate medicinal herbs for skincare via data mining a classical medical text. This enhances our understanding of such herbs and will help with discovering new candidate herbs.