Skip to content

Sentify

Financial Analysis

Leveraging Sentiment Analysis on News for Stock Market Insights.

Overview

This project focuses on analyzing the sentiment of news articles to predict stock trends. It features a Flask-based web application designed to provide real-time financial data analysis and sentiment tracking.

How It Works

Analysis Process

  • Input: Users can enter a company name or ticker symbol to retrieve relevant news articles.
  • Process: Each news article is analyzed for sentiment, with scores generated to gauge the article's impact on stock trends.
  • Output: The application delivers actionable insights and confidence indices based on the sentiment analysis, helping investors make informed decisions about their investments.

Features

Key Capabilities

  • Real-time Sentiment Analysis: Process news articles as they are published
  • Market Correlation: Link sentiment scores to market movements
  • Trading Signals: Generate buy/sell recommendations based on sentiment
  • News Aggregation: Collect and analyze news from multiple sources
  • Historical Analysis: Track sentiment trends over time
  • Web Interface: Flask-based application for easy access

Technology Stack

  • Backend: Flask web application
  • Language: Python 3.11+
  • Data Processing: Natural Language Processing (NLP)
  • News Sources: Financial news APIs
  • Analysis: Machine learning models for sentiment classification
  • Deployment: Local development server

Installation & Setup

Prerequisites

  • Python (version 3.11 or higher)
  • uv (recommended) or pip
git clone https://github.com/LifeAdventurer/sentify.git
cd sentify
uv sync
source .venv/bin/activate  # On Windows use `.venv\Scripts\activate`

Using pip

git clone https://github.com/LifeAdventurer/sentify.git
cd sentify
python -m venv venv
source venv/bin/activate  # On Windows use `venv\Scripts\activate`
pip install -r requirements.txt

Usage

To serve the Flask app locally, run:

python3 src/main.py

Configuration

The application can be configured by updating the config.py file in the src/config directory:

Key Configuration Parameters

  • TOP_COMPANIES_COUNT: Number of top companies ranked by market cap to search (default: 10000)
  • TIMESTAMP_FORMAT: Format for timestamps used in the application (default: "%Y-%m-%dT%H:%M:%SZ")
  • UTC_DIFFERENCE: Difference in hours between local time and UTC (default: 8)
  • MAX_NEWS_LOOKBACK_DAYS: Maximum days to look back when fetching news (default: 60)
  • CPU_COUNT: Number of CPUs used for multiprocessing (default: 2)

Use Cases

Perfect For

  • Individual Investors: Make data-driven investment decisions
  • Trading Firms: Enhance algorithmic trading strategies
  • Financial Analysts: Supplement traditional analysis methods
  • Market Researchers: Understand market sentiment trends
  • Risk Management: Identify potential market volatility

Repository

Source Code

View the project on GitHub

License

This project is licensed under the GNU General Public License v3.0 - see the LICENSE file for details.