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: can you tell me what types of movies you like? , USER: I like documentaries, biographies, and horror, like vintage horror movie British. , ASSISTANT: ok, why do you like those types of movies? , USER: I get a sense of nostalgia, and there isn't things flying through the air. And they use real people, real scenes, and no green screen. And there aren't like superhuman powers in the movies. Like I can relate to these movies better because they show real people doing real things, or they're in real houses. , ASSISTANT: got it, you like more realistic movies , USER: Yeah. Yes., classify_text: ok, can you name a film you really liked?
ASSISTANT