I am using neural network to detect spam in youtube comments and i have already used https://archive.ics.uci.edu/ml/datasets/YouTube+Spam+Collection data used for training but i want to retrieve more data since such dataset that i am using contains only 1956 comments which not enough for such a deep learning network.
I want to detect some way to extract such comments that have labeled classes to run over supervised learning classification and deep learning. If youtube doesn't provide such feature, can you provide me with a good sized dataset for training?
Comments
Post a Comment