0.0.13 Implementation Notes
Tagging process:
a. Partner provides base DCMI/MODS.
b. AI drafts Catholic tags (e.g., GPT‑4o with CCC lookup).
c. Editorial board reviews/approves (batch 50/week).
Recommendation matching (pseudo‑code):
score = dc_subject_similarity(profile.interests, content.subjects) * 0.4
+ match(formation_level, catholic:formation_level) * 0.3
+ match(season, catholic:spiritual_season) * 0.2
+ diversity_bonus(charism) * 0.1
Export/Interoperability: Full JSON‑LD validates against schema for partner sharing.
This schema is lean (under 30 fields total), standards‑compliant, and formation‑focused. Ready for v1 ingest and matching.
Respectfully submitted,
Michelle Tackabery
0.0.13 Implementation Notes
Tagging process:
a. Partner provides base DCMI/MODS.
b. AI drafts Catholic tags (e.g., GPT‑4o with CCC lookup).
c. Editorial board reviews/approves (batch 50/week).
Recommendation matching (pseudo‑code):
score = dc_subject_similarity(profile.interests, content.subjects) * 0.4
+ match(formation_level, catholic:formation_level) * 0.3
+ match(season, catholic:spiritual_season) * 0.2
+ diversity_bonus(charism) * 0.1
Export/Interoperability: Full JSON‑LD validates against schema for partner sharing.
This schema is lean (under 30 fields total), standards‑compliant, and formation‑focused. Ready for v1 ingest and matching.
Respectfully submitted,
Michelle Tackabery