Fraud Shield This Streamlit app allows users to identify potentially fraudulent financial news articles.
Features:
- Text input for entering news article or headline.
- Clickable button to initiate the search.
- Automatic analysis of the input text using OpenAI language model.
- Identification of potentially fraudulent content based on analysis.
- Differentiates between "Scam" and other types of content.
- Utilizes Google SERP tool for additional analysis.
Usage:
- Enter the text of the news article or headline in the provided text input field.
- Click the "Search" button to initiate the analysis.
- The app will provide feedback indicating whether the content is identified as a "Scam" or not.
Dependencies:
- Streamlit
- LangChain (including llms, agents, and tools modules)
- OpenAI API key (stored securely using Streamlit Secrets)
- Google SERP tool for additional analysis
Setup:
- Install the required dependencies.
- Obtain an OpenAI API key and store it securely in Streamlit Secrets.
- Ensure access to Google SERP tool.
- Run the Python script to start the Streamlit app.
Credits: This code was created by Raj Narayanan
Github: https://github.com/raju-2003
Explore and identify potentially fraudulent financial news articles with Fraud Shield!