Copy page

Core Concepts

Choosing Between Search, Natural Language, and Domain GenAI

The IntelCenter API provides three complementary ways to work with intelligence data:

• Structured Search — precise, filter-driven retrieval
• Natural Language Search — concept-based discovery
• Domain GenAI — synthesized answers and analysis

Understanding when to use each will significantly improve results and efficiency.

---

Overview Comparison

Capability	Structured Search	Natural Language Search	Domain GenAI
Input style	Keywords	Questions / phrases	Questions
Matching type	Exact / indexed	Conceptual	Reasoned synthesis
Output	Raw records	Ranked records	Generated answer
Best for precision	Excellent	Moderate	Low
Best for discovery	Moderate	Excellent	High
Best for answers	No	No	Yes
Uses AI reasoning	No	Yes (retrieval)	Yes (generation)

---

1. Structured Search

Endpoint examples:

• /v1/icd/videos
• /v1/icdb/images

Use When

• You know what you’re looking for
• You need exact filtering (entity, country, date)
• You want deterministic, repeatable results
• You are building structured applications or dashboards

Example

Code
/v1/icdb/videos?entity=Houthis [Ansar Allah]&date_from=2026-01-01

Strengths

• Fast and predictable
• Precise filtering
• Full control over results

Limitations

• Requires known keywords
• Misses conceptual or indirect matches

---

2. Natural Language Search

Endpoint examples:

• /v1/icd/nl-search
• /v1/icdb/nl-search

Use When

• You are exploring a topic
• You don’t know exact keywords
• You want conceptual matching across data types
• You need to discover relevant records

Example

Code
/v1/icd/nl-search?search=recent drone activity by Houthis [Ansar Allah]

Strengths

• Understands intent and context
• Finds matches across transcripts, visuals, and summaries
• Strong for discovery and investigation

Limitations

• Less precise than structured filters
• Returns results, not answers

---

3. Domain GenAI

Endpoint examples:

• /v1/icd/domain-genai
• /v1/icdb/domain-genai

Use When

• You need a direct answer or analysis
• You are asking complex or investigative questions
• You want summarized intelligence from multiple sources

Example

Code
curl --request POST \
  --url https://api.intelcenter.com/v1/icd/domain-genai \
  --header "Authorization: Bearer YOUR_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "question": "What recent operational trends are associated with Houthis [Ansar Allah]?"
  }'

Strengths

• Produces synthesized answers
• Combines multiple data sources
• Ideal for analysis and reporting

Limitations

• Higher cost and stricter rate limits
• Less control over exact source records
• Output is generated, not raw data

---

Recommended Workflow

For best results, combine all three approaches:

Step 1 — Discover

Use Natural Language Search to explore a topic

Step 2 — Refine

Use Structured Search to filter and validate results

Step 3 — Analyze

Use Domain GenAI to generate insights and summaries

---

Example Workflow

1. Start with Natural Language Search:

   "Drone usage by Houthis [Ansar Allah] in the past 6 months"

2. Review returned records and identify patterns

3. Use Structured Search:

   Code
   /v1/icdb/videos?entity=Houthis [Ansar Allah]&date_from=2025-10-01

4. Use Domain GenAI:

   "Summarize operational trends in drone usage by Houthis [Ansar Allah]"

---

Key Takeaways

• Structured Search = precision
• Natural Language Search = discovery
• Domain GenAI = understanding

Using them together provides the most complete intelligence workflow.

---

Tip

If you're unsure where to start:

• Start with Natural Language Search
• Move to Structured Search for control
• Use Domain GenAI for final insights