Given the task definition and input, reply with output. You are provided with a list of converations between a "USER" and "ASSISTANT". Figure out if the conversation provided as "classify_text" is authored by "USER" or "ASSISTANT". There are maximum of 7 conversations between "USER" and "ASSISTANT" in each input

ASSISTANT: What type of movies do you like? , USER: I really have discovered that I like horror movies a lot more than I thought I did. I'm a I'm a big fan of horror movies. , ASSISTANT: how did u discover that? , USER: Well, it just kind of dawned on me that I really enjoy when there is a good horror movie. So, I guess, you know? It's It has been kind of a long term in the works. There There're a lot of bad horror movies out there, so you know? It's hard for me to find a lot of good ones that I like, but once I have started to watch more, I realized that I'm really into them. , ASSISTANT: what is one of your top horror movies? , USER: One that I saw recently it was called The Witch. It's It was pretty interesting. I I hadn't seen a horror movie done like that before., classify_text: what else did u like about it?
ASSISTANT