Back to Knowledge Base

Dataset/Corpus Management

Filter and Search Dataset Items

Overview

Flow ID: filter-dataset-items
Category: Dataset/Corpus Management
Estimated Duration: 1-2 minutes
User Role: All Users
Complexity: Simple

Purpose: Search and filter dataset items to find specific information quickly. Search works across all text fields (block names, questions, answers) with real-time results.

Related Flows

Prerequisites

Before starting, users must have:

  • Dataset details page open
  • Dataset with items to search

Step-by-Step Flow

Main Path (Happy Path)

Step 1: Locate Search Box

  • User Action: On dataset details page, find search/filter input box
  • System Response: N/A
  • UI Elements Visible:
    • Filter input box (typically top-right of items area)
    • Placeholder text: "Search items..." or similar
    • Magnifying glass icon

Step 2: Enter Search Term

  • User Action: Click in search box and type search term
  • System Response:
    • Text appears as typed
    • After brief delay (300ms), filtering begins
    • Results update automatically
  • UI Elements Visible:
    • Search box with typed term
    • Loading briefly
    • Results list updates
    • Result count updates: "Showing 1-20 of 45 results"

Step 3: View Filtered Results

  • User Action: Review filtered items
  • System Response: Only matching items displayed
  • UI Elements Visible:
    • Items containing search term
    • Search term may be highlighted in results
    • Pagination resets to page 1
    • Page count reflects filtered results
  • Visual Cues:
    • Instant filtering (300ms delay)
    • Match highlighting possible

Step 4: Refine Search (Optional)

  • User Action: Modify search term to narrow or broaden results
  • System Response: Results update with each change
  • UI Elements Visible: Updated results matching new term

Step 5: Clear Search

  • User Action: Click X button in search box or delete all text
  • System Response:
    • Filter clears
    • All items shown again
    • Pagination resets
  • UI Elements Visible: Full dataset restored

Final Step: Found Desired Items

  • Success Indicator:
    • Relevant items displayed
    • Can browse filtered results
    • Search works reliably
  • System State Change:
    • View filtered to show matches
    • Can navigate and read filtered items
  • Next Possible Actions:
    • Read found items
    • Try different search terms
    • Clear search to view all
    • Page through results

Alternative Paths & Strategies

Strategy A: Multiple Keywords

When to use: Searching for combinations

Steps:

  1. Type multiple words separated by spaces
  2. System finds items containing all words
  3. Results show items matching all terms

Strategy B: Progressive Refinement

When to use: Too many results initially

Steps:

  1. Start with general term
  2. See large result count
  3. Add more specific terms
  4. Results narrow to most relevant

Error States & Recovery

Error 1: No Results Found

Cause: Search term doesn't match any items
User Experience:

  • Empty results
  • Message: "No items match search"

Recovery Steps:

  1. Check spelling
  2. Try different terms
  3. Use more general keywords
  4. Clear search to verify dataset has items

QA Note: Expected behavior, not error. Search working correctly.

Error 2: Search Very Slow

Cause: Very large dataset
User Experience:

  • Delay before results appear
  • May take several seconds

Recovery Steps:

  1. Wait for search to complete
  2. Results will appear
  3. Expected for 10,000+ item datasets

Version History

Date Version Author Changes
2025-10-04 1.1 Iternal Technologies Initial documentation

Notes

Search Behavior:

  • Searches across: block names, critical questions, trusted answers, source text
  • Case-insensitive
  • Partial word matching
  • Real-time with brief debounce (300ms)

Best Practices:

  • Use specific terms for targeted results
  • Try variations if first search doesn't find what expected
  • Use multiple keywords for precision

Common User Questions:

  • "Can I search multiple terms?" - Yes, space-separated finds items with all terms
  • "Is search case-sensitive?" - No
  • "Does it search all fields?" - Yes, across all item text content

Trigger

What initiates this flow:

  • User manually initiates

Specific trigger: User needs to find specific items in dataset rather than browsing all items.

User Intent Analysis

Primary Intent

Quickly locate specific information within large dataset by filtering items matching search criteria.

Secondary Intents

  • Verify dataset contains expected information
  • Find specific topics or keywords
  • Explore related items
  • Assess coverage of specific subjects
Free Trial

Download Blockify for your PC

Experience our 100% Local and Secure AI-powered chat application on your Windows PC

✓ 100% Local and Secure ✓ Windows 10/11 Support ✓ Requires GPU or Intel Ultra CPU
Start AirgapAI Free Trial
Free Trial

Try Blockify via API or Run it Yourself

Run a full powered version of Blockify via API or on your own AI Server, requires Intel Xeon or Intel/NVIDIA/AMD GPUs

✓ Cloud API or 100% Local ✓ Fine Tuned LLMs ✓ Immediate Value
Start Blockify API Free Trial