Query Loops

Query loops are a powerful feature of GeoRetina AI that allows you to build on previous query results, enabling multi-step analysis and progressive refinement of your insights.

What are Query Loops?

Query loops let you use the results of one query as the input for subsequent queries. This creates an iterative analysis workflow where you can:

• Narrow down from broad initial results to specific insights
• Apply multiple filters or conditions in sequence
• Build complex analytical workflows through simple conversational steps
• Refine your analysis based on intermediate findings

How Query Loops Work

When you run an initial query, GRAI returns a set of results. Rather than starting a completely new query, you can refer to "these results" or "the selected features" in your follow-up questions, creating a loop of increasingly refined analysis.

Example Workflow

1. Initial query: "Show me all schools in this region"

   

2. View initial results: GRAI returns matching features

   

3. Follow-up query: "Of these schools, show me only those with more than 500 students"

4. Further refinement: "Among these results, which ones are within 1km of a park?"

5. Final analysis: "For these remaining schools, what's the average distance to the nearest hospital?"

   

Each step builds on the previous result set, creating a logical chain of analysis.

Benefits of Query Loops

Query loops offer several advantages:

• Simplified complex analysis: Break down complex questions into manageable steps
• Interactive exploration: Adjust your approach based on intermediate findings
• Reduced data processing: Focus only on relevant subsets of data
• Intuitive workflow: Mirrors how people naturally think about multi-step analysis
• Transparent process: Each step is clearly documented in the conversation

Common Query Loop Patterns

Filter Refinement

Start with a broad set and progressively apply filters:

Comparative Analysis

Compare different subsets of data:

Spatial Relationships

Explore how features relate to each other:

Sequential Calculations

Perform calculations at each step:

Tips for Effective Query Loops

To get the most out of query loops:

• Start broad, then narrow: Begin with a larger dataset and refine
• Use clear references: Phrases like "these results," "these features," or "among these" help GRAI understand you're building on previous results
• Verify intermediate steps: Make sure each step gives you the expected results before proceeding
• Combine with other data sources: Bring in additional datasets at different stages of your analysis
• Save important results: Export critical findings at various points in your analysis workflow

Limitations

Some current limitations to be aware of:

• Query loops work best with vector data (as opposed to raster/satellite data)
• Very large result sets may need to be further filtered before detailed analysis
• Complex chains with many steps might require occasional clarification
• Some advanced spatial operations may not be available for all data types

Examples of Real-World Applications

Urban Planning

Environmental Analysis

Infrastructure Planning

Getting Started with Query Loops

The best way to master query loops is through practice. Start with simple two-step analyses and gradually build up to more complex workflows as you become familiar with the capabilities.