Data

Came up with some metrics for Farcaster profiles. If anybody has experience with ML or similar tech, I would be curious to hear ideas on how to feed casts and profile metadata into an algorithm that identifies spam.

Came up with some metrics for Farcaster profiles. If anybody has experience with ML or similar tech, I would be curious to hear ideas on how to feed casts and profile metadata into an algorithm that identifies spam. https://dune.com/queries/3540715

1/ Just shared with with an ML friend of mine - his feedback was that it's a pretty simple classification algo, so something like random forest or bayes could work, however some teams at Google have been having success with LLMs using the Gemma framework.

data | building | parenting | community | nonsense (in no particular order). "Likely not a bot"

2/ There's a codeset for fine-tuning it with LoRA here:

https://colab.research.google.com/github/google/generative-ai-docs/blob/main/site/en/gemma/docs/lora_tuning.ipynb

But you'll need a labelled dataset as a ground truth first