The aim of this project is to take txt or docx transcripts of interviews, then dynamically perform sentiment analysis on a predefined list of theme codes, where each of the interviewee's quotes are classified with the code. The quotes are defined dynamically, where it is defined after identifying the code. There will also be an explanation, ultimately resulting in three columns for a csv output.