to implant nlp_selector we need data for our model to train on,
training data should labeled by human, to reach a acceptable accuracy.
suggestion: we could create a website with an control panel to use crowdsourcing and invite people to tag persian tweets as an open source data, available for everyone.
to implant nlp_selector we need data for our model to train on, training data should labeled by human, to reach a acceptable accuracy. suggestion: we could create a website with an control panel to use crowdsourcing and invite people to tag persian tweets as an open source data, available for everyone.