Skip to content

Interactive Features Guide

Explore citation networks using the enhanced interactive Streamlit dashboard interface. Discover powerful new features including demo datasets, file upload, clickable network nodes, and real-time progress tracking.

Overview

Citation Compass provides an interactive web interface built with Streamlit. This guide covers the point-and-click features available through the dashboard, including recent enhancements for usability.

🚀 Getting Started

Launching the Dashboard

# Start the interactive dashboard
streamlit run app.py

The dashboard will open in your browser at http://localhost:8501 with a multi-page interface.

📊 Dashboard Pages

1. Home Page

Location: Main landing page

Features: - Enhanced platform overview with featured capabilities prominently displayed - Quick navigation cards to all major features with visual previews - System status indicators showing database, ML service, and analytics readiness - Getting started guidance with personalized recommendations

Home dashboard

2. 🎭 Demo Datasets

Location: Demo Datasets (in sidebar)

Features: - Instant data exploration - No setup required! - Curated academic datasets from multiple research fields - Dataset browser with expandable statistics (papers, citations, authors) - One-click loading with real-time progress indicators - Dataset comparison charts and performance metrics - Offline mode - Full functionality without database connection

Start Here!

Perfect for new users! Load the complete_demo dataset to explore all platform features with realistic academic data.

3. 📥 Data Import & File Upload

Location: Data Management → Data Import

Features: - Multiple import methods: Search queries, paper IDs, or file upload - Drag-and-drop file upload for .txt and .csv files with paper ID collections - Real-time progress tracking with streaming updates and performance metrics - Sample file downloads for testing and validation - Advanced configuration with quality filters and batch processing options - Error handling with detailed reporting and recovery options

4. ML Predictions

Location: Machine Learning → ML Predictions

Enhanced Features: - Demo mode support - Works with offline synthetic embeddings - Citation prediction interface with improved confidence visualization - Paper search with autocomplete and validation - Batch prediction capabilities for multiple papers - Interactive result exploration with sorting and filtering

ML predictions screen

Demo Mode Available

ML predictions work in demo mode using realistic synthetic embeddings. No trained model required for testing!

5. Embedding Explorer

Location: Machine Learning → Embedding Explorer

Enhanced Features: - Demo embeddings - Explore synthetic embeddings that cluster by research field - Interactive scatter plots with enhanced zoom and selection tools - Field-aware visualization - Papers cluster realistically by domain - Similarity exploration with confidence metrics

Embedding explorer

6. 🔗 Enhanced Visualizations with Clickable Nodes

Location: Analysis → Enhanced Visualizations

New Interactive Features: - Clickable network nodes - Click papers to view detailed information - Interactive citation paths - Trace relationships between papers - Dynamic filtering controls - Real-time network updates - Enhanced layouts - Improved force-directed and hierarchical arrangements - Performance optimizations - Smooth rendering for larger networks - Export capabilities with high-resolution outputs

Enhanced visualizations

7. Results Interpretation

Location: Analysis → Results Interpretation

Features: - Academic performance metrics with traffic-light indicators - Enhanced statistical interpretation with context and benchmarking - Comparison against academic standards - Improved report generation with LaTeX and PDF export options

8. Analysis Pipeline

Location: Analysis → Analysis Pipeline

Enhanced Features: - Interactive notebook execution with real-time progress tracking - Streaming parameter updates during analysis - Enhanced progress monitoring with detailed status information - Result visualization with improved interactivity

🎯 Key Interactive Features

🎭 Demo Mode Exploration

Instant Access: - No database required - Works without additional setup - Realistic academic data - Curated papers from AI, neuroscience, physics, and more - Full offline functionality - Complete feature access without internet - Educational workflows - Learn concepts with guided examples

Dataset Management: - Interactive dataset browser with expandable details - One-click dataset switching between different research domains - Performance monitoring - Load times, memory usage, processing speed - Comparison tools - Side-by-side dataset statistics

📁 File Upload & Import

Upload Interface: - Drag-and-drop functionality for .txt and .csv files - Real-time file validation with immediate feedback - Preview capabilities - See first 10 paper IDs before import - Sample file downloads - Get started with provided examples

Import Progress: - Streaming progress updates with real-time statistics - Performance metrics - Papers/second, success rates, error tracking - Status indicators - Visual progress bars and completion notifications - Error reporting - Detailed messages for troubleshooting

🔍 Enhanced Search and Discovery

Advanced Paper Search: - Intelligent autocomplete with search suggestions - Multi-field search by title, author, venue, or keywords - Advanced filtering by publication year, citation count, research field - Bulk operations - Select and process multiple papers simultaneously

Network Exploration with Clickable Nodes: - Interactive node clicking - Click any paper to view detailed information - Citation path tracing - Follow citation relationships visually - Dynamic zoom and pan with smooth animations - Real-time filtering - Update network display instantly - Enhanced highlighting - Emphasize important nodes and connections

🤖 ML Prediction Interface

Demo-Enhanced Predictions: - Offline prediction mode - Works with synthetic embeddings when no trained model available - Confidence visualization - Interactive confidence score displays - Field-aware results - Predictions consider research domain relationships - Temporal intelligence - Results reflect realistic academic citation patterns

Advanced Input Methods: - Smart paper ID validation - Real-time format checking - Title-based search with fuzzy matching - Batch prediction processing for multiple papers - Result comparison - Compare predictions across different papers

🎨 Visualization Controls

Interactive Network Elements: - Clickable nodes with hover information and detailed pop-ups - Dynamic filtering controls - Adjust network display in real-time - Color coding options - Research fields, publication years, citation counts - Layout algorithms - Force-directed, hierarchical, circular, and custom layouts - Animation controls - Smooth transitions and interactive animations

Enhanced Export Options: - High-resolution image export - PNG, SVG, PDF formats - Interactive HTML exports - Shareable network visualizations - Academic report generation - LaTeX tables and formatted documents - Data export formats - JSON, CSV, GraphML for further analysis

📊 Real-Time Analytics

Live Performance Monitoring: - Processing speed indicators - Real-time analysis performance - Memory usage tracking - Monitor system resource utilization - Progress streaming - Live updates during long-running operations - Error rate monitoring - Track and display operation success rates

Interactive Statistics: - Dynamic metric updates - Statistics change as you filter and explore - Comparative analysis - Benchmark against academic standards - Traffic-light indicators - Quick quality assessment with color coding - Trend visualization - See patterns emerge as you explore data

🔧 Customization Options

Dashboard Configuration

Most features can be customized through the interface:

  • Display preferences: Theme, layout, font sizes
  • Analysis parameters: Algorithm settings, thresholds
  • Visualization options: Colors, node sizes, edge styles
  • Export formats: File types, quality settings

Session Management

  • State persistence: Settings saved between sessions
  • Progress tracking: Analysis history and bookmarks
  • Data caching: Improved performance for repeated queries
  • Export history: Access previous exports

📈 Performance Tips

Optimizing Interactive Performance

  1. Start with Demo Mode: Use demo datasets to learn features without performance overhead
  2. Progressive Data Loading: Begin with smaller datasets before importing large collections
  3. Smart Caching: Enable caching for repeated analyses - works automatically
  4. Browser Optimization: Use Chrome or Firefox for best Streamlit performance
  5. Resource Management: Close unused browser tabs and applications
  6. Streaming Features: Take advantage of real-time progress updates for better UX

Memory Management for Large Networks

  • Use filtering controls to reduce displayed network size
  • Enable demo mode for resource-constrained environments
  • Monitor real-time metrics displayed in the interface
  • Leverage clickable nodes instead of displaying all details at once
  • Clear cache periodically using built-in cache management
  • Restart session if performance degrades (automatic session management available)

New Performance Features

  • Streaming pagination - Faster data loading for large imports
  • Intelligent batching - Automatic batch size optimization
  • Real-time progress - Live updates without blocking the interface
  • Offline capabilities - Full functionality without network dependencies

🛠️ Troubleshooting

Common Issues

Dashboard Won't Load: - ✅ Check that streamlit run app.py completed successfully - ✅ Verify port 8501 is available and not blocked by firewall - ✅ Try refreshing the browser and clearing cache - ✅ Check terminal for any startup errors or missing dependencies

Slow Dashboard Loading: - ✅ Try demo mode first - Loads instantly without database connection - ✅ Check system resources and close unnecessary applications - ✅ Verify database connection if using production mode - ✅ Use browser developer tools to check for JavaScript errors

Demo Datasets Not Loading: - ✅ Try loading smaller minimal_demo dataset first - ✅ Refresh the page and try again - ✅ Check browser console for error messages - ✅ Ensure adequate browser memory (close other tabs)

File Upload Failing: - ✅ Verify file format (.txt with IDs per line, or .csv with IDs in first column) - ✅ Check file encoding (should be UTF-8) - ✅ Try with provided sample files first - ✅ Ensure file size is under 200MB

ML Predictions Not Working: - ✅ Load demo dataset first - Provides synthetic embeddings - ✅ If using production mode, verify trained models exist in models/ directory - ✅ Check that ML service initialization completed successfully - ✅ Try with known paper IDs from loaded dataset

Slow Network Visualizations: - ✅ Use demo datasets for smooth performance testing - ✅ Reduce network size using filtering controls - ✅ Try different layout algorithms (some are faster) - ✅ Enable data sampling for very large networks

Import Progress Stalling: - ✅ Check internet connection stability - ✅ Verify API rate limits haven't been exceeded - ✅ Try reducing batch size in import configuration - ✅ Monitor system memory usage during import

New Troubleshooting Tools

  • Real-time status indicators - Check service health in the interface
  • Progress monitoring - See detailed import and processing status
  • Error reporting - Detailed error messages with suggestions
  • Demo mode fallback - Switch to offline mode if database issues occur

Getting Help

  • Check browser console for JavaScript error messages
  • Review Streamlit logs in the terminal for Python errors
  • Use demo mode to isolate issues from data/database problems
  • Visit GitHub Issues for community support
  • Check documentation for file upload guide and demo mode guide

🎨 Advanced Usage

Custom Visualizations

The dashboard supports custom visualization parameters:

  • Network layouts: Force-directed, circular, hierarchical
  • Color schemes: Categorical, continuous, custom palettes
  • Node sizing: By citation count, centrality, or custom metrics
  • Edge styling: Thickness, opacity, color coding

Integration with External Tools

  • Export compatibility: Gephi, Cytoscape, NetworkX formats
  • API endpoints: RESTful interface for external applications
  • Embedding integration: Compatible with TensorBoard, UMAP

📚 Next Steps

Ready to dive deeper? Explore these related guides:

  1. Demo Mode - Start here (no database required)
  2. Demo Datasets - Explore curated academic research data
  3. File Upload - Import your own research collections
  4. Data Import - Advanced import pipeline features
  1. Network Analysis - Advanced graph analysis features
  2. ML Predictions - Machine learning capabilities
  3. Results Interpretation - Understanding your results
  4. Notebook Pipeline - Programmatic analysis workflows
  1. Configuration - Database and API setup
  2. Developer Guide - Platform architecture
  3. API Reference - Programmatic interfaces
  4. Performance Optimization - Scaling and tuning

🎯 Quick Start Recommendations

For Researchers New to Citation Analysis

Start with: Demo Mode → Load complete_demo → Explore Enhanced Visualizations → Try ML Predictions

For Data Scientists

Start with: Demo Datasets → API exploration → Notebook Pipeline → Custom model training

For Research Administrators

Start with: Demo Mode → File Upload → Data Import → Scale planning

Happy exploring with enhanced interactive features! 🚀✨