CosFunc-HO: Hierarchical Cosmetic Ingredient Function Dataset

CosFunc-HO is a cleaned and structured cosmetic ingredient function dataset developed from COSDNA data. It contains more than 9,000 labeled ingredient records with hierarchical functional annotations. The dataset was created by hamdanzameer and was last updated on 2026-05-22.

Use Cases

• Multi-label classification of cosmetic ingredient functions based on hierarchical annotations.
• Ontology learning for cosmetic chemistry based on structured ingredient-function relationships.
• Training and evaluating Large Language Models (LLMs) on domain-specific, structured scientific data.
• Analyzing ingredient prevalence and functional categories within cosmetic products.

Strengths

• Contains more than 9,000 labeled cosmetic ingredient records.
• Includes hierarchical functional annotations organized into multiple levels of abstraction.
• Dataset is cleaned and structured from a known source (COSDNA).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last updated 2026-05-22 12:40:07; freshness should be verified.

Provenance

Source

COSDNA data

Collection Method

Developed, cleaned, and structured from source data.

Freshness

2026-05-22