Understanding Dietary Recall Auto-Matching
Nutritional Matching
•
3 Views
•
May 19, 2026
FoodQuant utilizes double-layered lexical matching engines to align raw respondent text descriptions to verified food composition tables.
How the Matcher Operates
Lexical Normalization: Special characters, weights, and comments are filtered out.Similarity Scoring: FoodQuant measures edit distances (Levenshtein) and trigram token matches against database tables (e.g., USDA FoodData Central and regional tables).Score Thresholds: Perfect Match (>= 90%): Mapped automatically. Fuzzy Match (60% - 89%): Flagged for researcher approval. No Match (< 60%): Marked as unmatched, prompting you to search manually or map to a custom My Mix reference item.
Was this article helpful?
Help us improve the Help Desk for researchers worldwide.
