What is the best methods to detect harmful content such as racial abuse in tweets?

I’m thinking about a research project in which I try various methods and compare their accuracy. Am I right in thinking that Naive Bayes, Logistic Regression, Support Vector Machine, LSTM and BERT would be some of the best methods?

  • tail-recursion@alien.topB
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 year ago

    You could try an open source LLM like Llama 2. You could probably use Langchain tools to give it a tool to tag when a tweet has harmful content.