Document Analysis¶

Learn about Archivus's AI analysis features including summarization, entity extraction, and classification.

Overview¶

Archivus provides comprehensive AI analysis of your documents:

• Summarization - Quick and full summaries
• Entity Extraction - People, organizations, dates, amounts
• Classification - Automatic categorization
• Duplicate Detection - Find similar documents

Summarization¶

Quick Summary¶

Get a 2-3 sentence overview:

1. Open document
2. Click "Summary" tab
3. View AI-generated quick summary

Example:

"This document is a service agreement between Acme Corp and XYZ Inc, covering a 24-month term with payment terms of Net 30 days."

Full Summary¶

Get a comprehensive summary:

1. Open document
2. Click "Generate Full Summary"
3. View detailed summary with:
4. Key points
5. Important dates
6. Parties involved
7. Terms and conditions

Entity Extraction¶

Automatically extract key information:

People¶

Names and roles mentioned in document:

• John Doe (CEO)
• Jane Smith (CFO)

Organizations¶

Companies and institutions:

• Acme Corp (Service Provider)
• XYZ Inc (Client)

Dates¶

Important dates and deadlines:

• Start Date: January 1, 2025
• End Date: December 31, 2026
• Payment Due: 30 days from invoice

Amounts¶

Financial figures and prices:

• Monthly Payment: $50,000
• Total Value: $1,200,000

Locations¶

Places mentioned:

• New York, NY
• San Francisco, CA

Classification¶

Automatic document categorization:

Document Types¶

AI identifies document type:

• Contract
• Invoice
• Report
• Agreement
• Policy
• Receipt

Categories¶

High-level categories:

• Legal
• Financial
• Business
• HR
• Compliance

Viewing Classification¶

1. Open document
2. See "Type" and "Categories" tags
3. AI confidence shown for each

Duplicate Detection¶

Find similar or duplicate documents:

1. Open document
2. Click "Find Similar"
3. See list of similar documents with:
4. Similarity score (0.0-1.0)
5. Reason for match
6. Quick preview

Use cases:

• Find duplicate uploads
• Discover related documents
• Check for similar contracts
• Find previous versions

How to Access Features¶

From Document View¶

1. Open any document
2. Click on tabs:
3. Summary - View summaries
4. Entities - See extracted entities
5. Classification - View categories
6. Similar - Find related documents

Bulk Analysis¶

Analyze multiple documents:

1. Select documents (Cmd/Ctrl+click)
2. Click "Bulk Actions"
3. Select analysis action
4. Processing begins for all

Best Practices¶

Summarization¶

• Use quick summaries - For overviews and lists
• Use full summaries - For detailed analysis
• Regenerate if needed - AI improves with document updates

Entity Extraction¶

• Review confidence scores - Verify low-confidence entities
• Use for automation - Extract data for workflows
• Combine with tags - Use entities as tags

Classification¶

• Review categories - Verify AI classification
• Use for organization - Organize by category
• Correct if wrong - AI learns from corrections

Duplicate Detection¶

• Set threshold - Adjust similarity threshold (default 0.8)
• Check before deleting - Verify duplicates before removal
• Use for cleanup - Regular duplicate checks

AI Learning¶

AI improves over time:

• Corrections - When you correct AI, it learns
• Preferences - AI learns your tagging preferences
• Patterns - Recognizes patterns in your documents
• Accuracy - Improves with more documents

Troubleshooting¶

Wrong Summary¶

If summary is inaccurate:

1. Regenerate - Click "Regenerate Summary"
2. Check document - Ensure fully processed
3. Try full summary - More comprehensive
4. Report issue - Help AI improve

Missing Entities¶

If entities not found:

1. Check processing - Document must be fully processed
2. Regenerate - Click "Extract Entities" again
3. Manual entry - Add entities manually if needed

Wrong Classification¶

If classification is wrong:

1. Correct it - Update document type/category
2. AI learns - Your correction helps AI improve
3. Add tags - Manual tags supplement AI

Next Steps¶

• Chat with Documents - Ask questions
• Auto-Tagging - Automatic organization