| Stage | Rule | In | Kept | Removed |
|---|---|---|---|---|
| Google search | Daily scrape across 13 search terms; links from YouTube, TikTok, Facebook, Instagram | — | 18,579 | — |
| Download + LLM | Video retrieved by yt-dlp and processed by Nemotron 3 Nano Omni; drops unretrievable links (most Instagram), audio-only files and processing failures | 18,579 | 12,213 | 6,366 |
| Menopause screen | LLM field menopause = True (post targets the supplement toward menopause-related symptoms) | 12,213 | 4,447 | 7,766 |
| Deduplication | Same normalised title (≥ 15 chars) and duration across platforms; copy with most views kept | 4,447 | 4,338 | 109 |
| Supplement screen | Post names at least one supplement; the 1,404 without are retained separately as symptom-and-experience posts | 4,338 | 2,934 | 1,404 |
| Primary category | Modal status category of the post's supplement terms, per the team's nine-category coding framework; unmatched terms go to the manual review queue | 2,934 | 2,934 | — |
| From | To | Posts | % of source |
|---|---|---|---|
| Downloaded & LLM-processed | 151 | 2.6% | |
| Not retrievable / not processed | 5,548 | 97.4% | |
| TikTok | Downloaded & LLM-processed | 4,819 | 93.0% |
| TikTok | Not retrievable / not processed | 363 | 7.0% |
| YouTube | Downloaded & LLM-processed | 3,845 | 97.9% |
| YouTube | Not retrievable / not processed | 84 | 2.1% |
| Downloaded & LLM-processed | 3,398 | 90.2% | |
| Not retrievable / not processed | 371 | 9.8% | |
| Downloaded & LLM-processed | Menopause-targeted | 4,447 | 36.4% |
| Downloaded & LLM-processed | Not menopause-targeted | 7,766 | 63.6% |
| Menopause-targeted | Unique posts | 4,338 | 97.5% |
| Menopause-targeted | Cross-platform duplicates | 109 | 2.5% |
| Unique posts | Mentions supplements | 2,934 | 67.6% |
| Unique posts | No supplements mentioned | 1,404 | 32.4% |
| Mentions supplements | Manual review queue | 838 | 28.6% |
| Mentions supplements | Vitamins & minerals | 759 | 25.9% |
| Mentions supplements | Branded menopause formulations | 457 | 15.6% |
| Mentions supplements | Hormone therapies | 263 | 9.0% |
| Mentions supplements | Protein, amino acids & creatine | 202 | 6.9% |
| Mentions supplements | Herbals & botanicals | 186 | 6.3% |
| Mentions supplements | Gut / microbiome | 72 | 2.5% |
| Mentions supplements | Vague or non-classifiable | 67 | 2.3% |
| Mentions supplements | Fatty acids & oils | 57 | 1.9% |
| Mentions supplements | Prescription & OTC medicines | 33 | 1.1% |
Bands are proportional to post counts; nodes smaller than 117 posts are drawn at a 4px minimum so they stay visible. Category assignment gives each post the most frequent status category among its supplement terms (ties resolved away from the manual-review flag). Generated by src/generate_sankey.py from data/supplements.csv (search links), data/supplements_LLM_results.xlsx (LLM output), src/clean_supplements.py (category coding) and src/analyze_data.py (platform filter and deduplication).