I have a dream of making a cringe detector - a program that you could input a tweet or image, and it outputs how cringy it is.<p>I am thinking, it could use a neural net, or not necessary.<p>I was trying to imagine, how could that thing work? And actually out of clue.<p>What do you think?
look up “sentiment analysis NLP”. That is what you want, except instead of detecting whether text has a positive or negative opinion, you detect whether it’s “cringe” / “not cringe”.<p>I would start by building a sentiment-analysis model by following a tutorial like [1]. Except instead of a dataset with data labeled “positive opinion” and “negative opinion”, you need data labeled “cringe” and “not cringe”. The hard part is finding and classifying that data: remember that if you just take a bunch of “cringe” and ordinary tweets, you may be training for spurious correlations like “hot take” which your model will misinterpret as being “cringe”<p>[1] <a href="https://huggingface.co/blog/sentiment-analysis-python#3-building-your-own-sentiment-analysis-model" rel="nofollow">https://huggingface.co/blog/sentiment-analysis-python#3-buil...</a>
if trained via reddit, i think you could call any content cringe. maybe by finding people from different segments to rate whether certain content is cringe.