Defame or Nah

"Defame or Nah" aims to create an innovative online tool that empowers individuals to determine whether a publicly published message is defamatory or not, providing them with clarity and confidence in their reputation management.

Problem Statement: In today's digital age, publicly published messages can have a significant impact on one's reputation. With the rise of social media and online content, verifying whether a statement is defamatory has become increasingly challenging. Existing solutions often rely on manual research or consultation with lawyers, which can be time-consuming and costly.

Background: Defamation laws can be very complicated, making it essential to develop a platform that can accurately identify potential defamation cases. Traditional methods of determining defamation, such as reviewing public court sentences and precedents, are resource-intensive and not easily scalable.

Stakeholders and Their Needs:

1. Individuals: They need an easy-to-use tool to determine whether a publicly published message is defamatory or not so they can take legal action if needed
2. Law Agencies: They require a reliable platform for investigating defamation cases and gathering evidence to protect the fame of their client
3. Celebrities: They need to protect their reputation from defamation.

Vision of Solution: The "Defame or Nah" platform aims to provide tool that:

1. Allows users to input a publicly published message.
2. Uses web scraping and natural language processing techniques to analyze the message for potential defamation.
3. Cross-references public court sentences and precedents using secondary data analysis to determine the likelihood of defamation.

Features:

1. User-friendly interface for inputting messages.
2. Natural language processing (NLP) algorithms to analyze message sentiment and identify potential defamation.

Quick start

Using Docker

$ docker build -t defameornah .
$ docker run --name defameornah -p 8000:8000 -d defameornah

You can now visit via localhost:8000

Manually

1. Install Ollama
2. Pull bge-m3 model using ollama
3. Clone this repo
4. Create or activate virtual environment (optional)
5. pip install -r requirements.txt
6. Generate stub: fastapi-codegen --input .openapi/api.yml --output ./stub --output-model-type pydantic_v2.BaseModel --template-dir ./templates
7. Run: fastapi run main.py

You can now visit via localhost:8000