Durian_knowledge_base / docs /data_dictionary.md
mthegn
feat: Initial commit v1.0.0 (Clean LFS)
d204742

📖 Durian Knowledge Base - Data Dictionary

This document explains the meaning of each key in the dataset.json file (as of v1.0.0).


1. Root Object

The JSON file is a single object with three main keys:

  • "metadata": (Object) General information about the dataset.
  • "diseases": (Array) A list of pathological diseases (caused by fungi, algae, etc.).
  • "pests": (Array) A list of harmful pests (insects, mites, nematodes).

2. Metadata Fields

Contains descriptive information about the dataset itself.

  • "name": The full name of the dataset.
  • "version": The current version (following SemVer MAJOR.MINOR.PATCH).
  • "description": A brief description of the dataset's purpose.
  • "creator": (Object) Information about the author.
    • "name", "affiliation", "email", "orcid"
  • "created_date": The date the project was initiated.
  • "last_updated": The date the JSON file was last modified.
  • "license": The license code (e.g., "CC BY 4.0").
  • "geographical_coverage": The geographical region this data applies to (e.g., "Mekong Delta", "Tây Nguyên").
  • "number_of_diseases": Total count of items in the diseases array.
  • "number_of_pests": Total count of items in the pests array.
  • "data_source": A general description of the data sources.
  • "repository_url": The URL to this GitHub repository.

3. Disease and Pest Fields

Both diseases and pests are arrays of objects. Each object describes one specific problem.

3.1. General Information

  • "disease_id" (or "pest_id"): A unique identifier (e.g., "SR001", "P001").
  • "disease_name" (or "pest_name"): The most common name (e.g., "Stem Canker").
  • "scientific_name": The scientific name of the pathogen or pest (e.g., "Phytophthora sp.").
  • "common_names": (Array) Other common names.
  • "pathogen_type" (or "pest_type"): The classification of the agent (e.g., "nấm" (fungus), "tảo" (algae), "Côn trùng chích hút" (Sucking insect)).
  • "severity_level": The assessed danger level.
    • "rất_cao" (Very High): Can kill the tree or cause total crop loss.
    • "cao" (High): Causes severe damage and significant yield reduction.
    • "trung_bình" (Medium): Causes harm but is less likely to be fatal to the tree.
  • "risk_season": (Array) Seasonal conditions when outbreaks are most likely (e.g., "mùa mưa" (rainy season), "mùa khô" (dry season)).

3.2. Symptoms (symptoms)

An object describing symptoms on different plant parts.

  • "leaves", "stem", "roots", "fruits", "flowers": (Array) An array of strings describing symptoms.
  • "early_stage": (String) Description of the earliest detectable signs.
  • "late_stage": (String) Description of the consequences of a severe infestation.

3.3. Conditions & Prevention

  • "favorable_conditions": (Array) Environmental or cultural conditions that favor the problem.
  • "prevention_methods": (Object) Proactive prevention measures.
    • "cultural": (Array) Farming practices (pruning, fertilizing, sanitation...).
    • "biological": (Array) Biological controls (natural enemies, beneficial fungi...).
    • "chemical_prevention": (Array) Chemical prophylaxis (painting with lime, preventative spraying...).

3.4. Treatment Chemicals (treatment_chemicals)

An array of recommended pesticides/fungicides.

  • "chemical_id": A unique ID for the chemical entry (e.g., "SR001-CHEM01").
  • "name": A common trade name (e.g., "Ridomil Gold 68WG").
  • "active_ingredient": The chemical's active ingredient (e.g., "Metalaxyl + Mancozeb").
  • "type": The mechanism of action.
    • "tiếp xúc" (Contact): Kills on contact.
    • "lưu dẫn" (Systemic/Internal): Absorbed and translocated by the plant.
    • "vị độc" (Stomach Poison): Must be ingested by the pest.
    • "xông hơi" (Fumigant): Kills via gas/vapor.
  • "application_method": (String) How to apply it (e.g., "Phun lá" (Foliar spray), "Tưới gốc" (Soil drench)...).
  • "source": (String) The source that recommended this specific chemical.

3.5. Other Fields

  • "diagnosis_tips": (Array) Key "tricks" for accurate identification (e.g., "Dig up roots to check", "Wipe the leaf...").
  • "verified_sources": (Array) A list of official sources used to compile the data.
  • "images": (Array) A list of illustrative image objects.
    • "image_id", "description", "filename".
  • "expert_notes": (String) A critical summary or "most important thing to know".
  • "data_quality": (Object) The creator's self-assessment of the data.
    • "completeness": How complete is the data for this entry (0.0 - 1.0).
    • "accuracy": How accurate is the data (0.0 - 1.0).
    • "consistency": How consistent is the data across sources (0.0 - 1.0).
    • "last_verified": The date this information was last fact-checked.