My research domain pertains to Deep Learning, and with reviews. To make better feature selection and extraction by deep learning algorithm I have a collection of close to a million review data. But, this is data is very unclean. I want to know if there are any tools or fast-track techniques through which I can clean my data automatically. Also, if the answer is no then is there any faster way to clean the data manually? Please lend me your expert advice in this regard.