Skip to content

dashboard-guide.md

Latest commit

 

History

History
117 lines (90 loc) · 2.84 KB

dashboard-guide.md

File metadata and controls

117 lines (90 loc) · 2.84 KB

Visual Dashboard Guide

Complete guide to using the interactive UOM Intelligence Dashboard.

Overview

The Visual Dashboard is the heart of the platform - a real-time, interactive interface that shows:

  • Error detection in progress
  • Confidence scores and classifications
  • Auto-correction results
  • Statistical analytics
  • Color-coded visualizations

Accessing the Dashboard

  1. Execute the KNIME workflow
  2. Right-click the "Visual Dashboard" node
  3. Select "Open View"
  4. Dashboard opens in a new window

Dashboard Components

Summary Statistics

  • Total Records processed
  • Errors Found (count and percentage)
  • Auto-Corrected (count and percentage)
  • Manual Review needed (count and percentage)

Error Classification Chart

Visual breakdown of error types:

  • Decimal Errors (35%)
  • Unit Mismatches (30%)
  • Conversion Issues (20%)
  • Format Problems (15%)

Confidence Score Distribution

Histogram showing ML model confidence:

  • 90-100%: High confidence (auto-correct)
  • 70-89%: Medium confidence (review suggested)
  • Below 70%: Low confidence (manual required)

Detailed Results Table

Scrollable table with individual records showing:

  • Record ID
  • Original UOM
  • Detected Error Type
  • Suggested Fix
  • Confidence Score
  • Status (Auto-corrected/Manual Review)

Interactive Features

Filtering and Sorting

  • Click column headers to sort
  • Filter by error type
  • Search specific records
  • Export results (CSV, Excel, PDF)

Real-Time Updates

As processing continues:

  • Charts update automatically
  • Progress bar advances
  • New errors appear instantly

Color Codes

Status Colors

  • Green: Valid/Corrected
  • Yellow: Needs review
  • Red: Critical error
  • Gray: Processing

Confidence Colors

  • Green (90%+): Very confident
  • Yellow (70-89%): Moderately confident
  • Red (<70%): Low confidence

Best Practices

  1. Monitor during processing to identify patterns
  2. Review medium-confidence items (70-89%)
  3. Export results regularly for historical tracking
  4. Use filters to focus on specific error types
  5. Document patterns for process improvement

Troubleshooting

Dashboard Not Loading

  • Check JavaScript is enabled in KNIME
  • Verify workflow executed successfully
  • Try refreshing the view

Slow Performance

  • Reduce table row limit
  • Disable real-time updates
  • Close other KNIME views

Missing Data

  • Ensure workflow completed
  • Check input data loaded correctly
  • Re-execute the workflow

Advanced Features

Export Options

  • CSV: For Excel/database import
  • Excel: With formatting
  • PDF Report: For documentation

Custom Views

  1. Click Settings (gear icon)
  2. Adjust chart types, colors, columns
  3. Save layout for reuse

Questions? Open an issue or check the FAQ