Getting Started

Welcome to CheatInSights! This platform helps you analyze Microsoft Word documents for potential academic integrity issues by examining document metadata, RSID patterns, and cross-document similarities.

What is CheatInSights?
CheatInSights is an advanced document analysis tool that uses sophisticated algorithms to detect potential plagiarism, collusion, and other academic integrity violations in Microsoft Word documents.

Quick Start Guide

  1. Upload one or more .docx files using the upload interface
  2. Wait for the analysis to complete (usually takes a few seconds)
  3. Review the suspicion scores and detailed analysis results
  4. Examine RSID patterns and cross-document similarities
  5. Use the interactive features to explore document content

Features

📊 Suspicion Scoring

Advanced algorithms calculate suspicion scores based on multiple factors including metadata inconsistencies, RSID patterns, and document anomalies.

🔍 RSID Analysis

Deep analysis of Revision Session IDs to detect copy-paste operations and identify unusual editing patterns.

📋 Cross-Document Analysis

Compare multiple documents to identify shared authors, RSIDs, and potential collusion between students.

📈 Statistical Analysis

Comprehensive statistical analysis including character counts, word counts, and paragraph analysis.

🎨 Interactive Visualization

Interactive charts and graphs to visualize document patterns and statistical distributions.

📱 Responsive Interface

Modern, responsive web interface that works on desktop, tablet, and mobile devices.

Uploading Documents

CheatInSights supports Microsoft Word documents (.docx format) for analysis. Here's how to upload your documents:

Supported File Formats

  • .docx - Microsoft Word documents (recommended)
  • Multiple files - Upload up to 1000 documents simultaneously
Important: Only .docx files are supported. .doc files (older Word format) are not compatible.

Upload Process

  1. Navigate to the upload page or dashboard
  2. Click "Choose Files" or drag and drop files into the upload area
  3. Select one or more .docx files from your computer
  4. Click "Upload" to begin the analysis
  5. Wait for processing to complete (progress bar will show status)

File Size Limits

  • Maximum file size: 50MB per file
  • Maximum total upload: 500MB
  • Recommended: Keep files under 10MB for faster processing

Understanding Results

After uploading documents, CheatInSights provides comprehensive analysis results across multiple views:

Dashboard Overview

The main dashboard shows:

  • File List: All uploaded documents with metadata and suspicion scores
  • Suspicion Scores: Percentage-based scores indicating potential issues
  • Document Statistics: Character counts, word counts, and other metrics
  • Cross-Document Analysis: Shared authors and RSIDs between documents

Document Views

View Purpose Features
Files Overview of all documents Metadata, scores, search, filter
Focus Detailed single document analysis RSID highlighting, metadata, reconstruction
Graphs Statistical visualizations Charts, distributions, comparisons

Suspicion Scoring

CheatInSights uses a sophisticated scoring system to identify potential academic integrity issues:

Scoring Factors

Factor Weight Description
Different Author/Modifier 15 points Author and last modified by are different people
Modified Before Created 25 points Document was modified before its creation date
Missing Metadata 15 points Key metadata fields are missing or empty
Long Run Outliers 25 points Unusually long text runs suggesting copy-paste
Author Collusion 30 points Same author appears in multiple documents
RSID Collusion 30 points Same RSIDs appear in multiple documents

Score Interpretation

0-30% (Low Risk)
Document appears to be original work with minimal suspicious indicators.
31-70% (Medium Risk)
Some suspicious indicators detected. Review recommended.
71-100% (High Risk)
Multiple suspicious indicators detected. Thorough investigation recommended.

RSID Analysis

Revision Session IDs (RSIDs) are unique identifiers that Microsoft Word assigns to text runs. They provide valuable insights into document creation patterns:

What are RSIDs?

RSIDs are 32-character hexadecimal strings that identify when and how text was created or modified. Each text run in a Word document has an associated RSID.

RSID Analysis Features

  • RSID Highlighting: Visually highlight all text with the same RSID
  • RSID Statistics: Count and analyze RSID patterns
  • Cross-Document RSID Matching: Find shared RSIDs between documents
  • RSID Timeline Analysis: Understand document creation sequence

Interpreting RSID Patterns

Normal Pattern: Multiple RSIDs with reasonable character distributions
Suspicious Pattern: Very few RSIDs with large character counts (suggesting copy-paste)

Cross-Document Analysis

When multiple documents are uploaded, CheatInSights performs cross-document analysis to detect potential collusion:

Analysis Types

  • Author Analysis: Identifies documents with the same author
  • RSID Analysis: Finds shared RSIDs between documents
  • Metadata Comparison: Compares creation dates, modification patterns
  • Content Similarity: Analyzes text patterns and structures

Collusion Detection

The system flags potential collusion when:

  • Same author appears in multiple documents
  • Same RSIDs are found across different documents
  • Documents have similar creation/modification patterns
  • Statistical patterns suggest shared content

Troubleshooting

Common issues and solutions:

Upload Issues

Problem: "Only .docx files are allowed"
Solution: Convert your .doc file to .docx format using Microsoft Word
Problem: "File too large"
Solution: Reduce file size by removing images or splitting into smaller documents

Analysis Issues

Problem: "No RSIDs found"
Solution: This is normal for very simple documents or documents created in older Word versions
Problem: "Low suspicion score despite concerns"
Solution: Review the detailed factors and consider manual inspection

Performance Issues

  • Large files may take longer to process
  • Multiple files are processed sequentially
  • Browser cache may need clearing for optimal performance