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 kind of movies do you like? , USER: I like movies with an art style, good architecture, good music, a unique story line. , ASSISTANT: ok , ASSISTANT: can you think of a movie that had that? , USER: Yes, I can. Grand Budapest Hotel, Inception , ASSISTANT: ok, Inception, what about that kind of movie do you recall liking?, classify_text: I liked the music, the atmosphere, the architecture, and the story line.
USER