Commit Graph
7 Commits
Author SHA1 Message Date
KaifAhmad1 04c4c9fb4c docs: clean and fix corrupted notebooks in cookbook 2026-01-07 15:41:13 +05:30
KaifAhmad1 9c97d3236c Fix Fraud Detection notebook: Add real data sources, fix errors, enhance GraphRAG with Context Graph
- Add real CSV and JSON data sources for transactions and accounts
- Fix ConflictDetector, TemporalGraphQuery, and Reasoner errors
- Simplify code to use Semantica modules properly
- Enhance GraphRAG section with Context Graph and Groq LLM
- Add temporal interactive visualization using TemporalVisualizer
- Fix CSV export to use CSVExporter instead of GraphExporter
- Update README.md to mention Context Graph and Context Retriever
2025-12-30 13:45:56 +05:30
KaifAhmad1 a674e8c039 fix: resolve Entity TypeError by adding required start_char and end_char fields across cookbook notebooks 2025-12-29 21:35:30 +05:30
KaifAhmad1 8289d56d89 Update notebook and other changes 2025-12-26 19:41:55 +05:30
KaifAhmad1 c15ee40cdf feat: Diversify deduplication and conflict resolution across 16 use case notebooks
Implement domain-appropriate strategies for all notebooks:

Deduplication Methods (9): pairwise, batch, incremental, group, graph_based,
hierarchical, exact, semantic, fuzzy

Merge Strategies (5): keep_first, keep_last, keep_most_complete,
keep_highest_confidence, merge_all

Conflict Detection (6): value, type, entity, relationship, temporal, logical

Conflict Resolution (6): voting, credibility_weighted, most_recent,
first_seen, highest_confidence, expert_review

Key patterns:
- Real-time: pairwise + keep_first + first_seen
- Time-sensitive: temporal + most_recent
- Multi-source: batch + merge_all + voting
- Medical/Research: credibility_weighted
- Fraud/Security: graph_based + logical + expert_review

Added STRATEGIES_SUMMARY.md documentation.
Removed temporary update scripts.
2025-12-25 22:38:03 +05:30
KaifAhmad1 7d07691d99 Refactor finance notebooks: Use modular architecture with unique approaches per use case
- Rebuilt 01_Financial_Data_Integration_MCP.ipynb with MCP and seed data focus (31 cells)
- Rebuilt 02_Fraud_Detection.ipynb with temporal analysis and pattern detection focus (38 cells)
- Added comprehensive data sources (APIs, RSS feeds, streams, databases)
- Implemented cell-specific imports and minimal print statements
- Each notebook uses different module combinations to showcase uniqueness:
  - Financial Data Integration: MCPIngestor, SeedDataManager, GraphAnalyzer, CentralityCalculator
  - Fraud Detection: StreamIngestor, TemporalGraphQuery, TemporalPatternDetector, ConflictDetector
- Removed empty markdown cells
- Added Mermaid pipeline flow diagrams
- No phase/step numbers - descriptive section headers
2025-12-25 14:12:27 +05:30
KaifAhmad1 106e817ad8 docs: Update all documentation with 18 enhanced domain-specific cookbooks
- Enhanced 18 cookbooks across 9 domains with real data sources, advanced chunking, temporal KGs, and GraphRAG
- Updated docs/cookbook.md with all 18 cookbook links and enhanced descriptions
- Updated docs/use-cases.md with corrected links and removed duplicates
- Updated README.md with comprehensive Industry Use Cases section
- Fixed all outdated notebook links and ensured consistency across all docs
- Added real data ingestion (RSS feeds, APIs, MCP servers, streams)
- Integrated advanced chunking strategies (entity-aware, relation-aware, ontology-aware, semantic_transformer, etc.)
- Added temporal knowledge graphs, GraphRAG, deduplication, conflict detection, and other Semantica modules
2025-12-24 16:33:36 +05:30