FinnCorr - Financial Data Correlation Analysis Engine

Overview

A REST API and web-based analysis tool for financial time series correlation analysis, featuring automated data processing and visualization capabilities.

Features

Data Processing

• CSV and JSON file support
• Automated price data detection
• Multiple data format handling
• Time series data support

Analysis Capabilities

• Percentage change correlation calculations
• Pearson correlation coefficient analysis
• Correlation classification:
  • Strong positive (> 70%)
  • Moderate positive (30-70%)
  • Little/no correlation (-30% to 30%)
  • Moderate negative (-30% to -70%)
  • Strong negative (< -70%)

Visualization

• Interactive Plotly charts
• Dual-axis correlation visualization
• HTML-based output

Technical Stack

• ASP.NET Core REST API
• ML.NET for data processing
• SoftCircuits.CsvParser
• System.Text.Json
• Plotly.js

API Reference

Upload Endpoint

POST /api/upload
Content-Type: multipart/form-data

Response Structure

interface AnalysisResult {
    insights: string;
    graphUrl: string;
    totalCorrelations: number;
    strongPositive: number;
    moderatePositive: number;
    moderateNegative: number;
    strongNegative: number;
    noCorrelation: number;
    positivePercentage: number;
    negativePercentage: number;
}

Implementation Details

Core Components

• DataAnalysisService: Analysis logic
• UploadController: API handling
• FileUploadDto: Upload model
• AnalysisResult: Output model
• AnalysisConfiguration: Settings model

Mathematical Foundation

Percentage Change

Δ% = ((pₙ - pₙ₋₁) / pₙ₋₁) × 100

Pearson Correlation

r = Σ((x - μₓ)(y - μᵧ)) / √(Σ(x - μₓ)² × Σ(y - μᵧ)²)

Quality Scoring

Q = (C × 10) + V + R + S

Setup

1. Install .NET runtime
2. Clone repository
3. Run dotnet restore
4. Execute dotnet run
5. Access Swagger UI at root URL

Documentation

Full API documentation available via Swagger UI