Predictors for Movie Box Office Success
Introduction:
This project looks at the public movie metadata (from Tmdb) and user ratings (MovieLens | GroupLens) to analyze predictors of the movie revenue.
Source Data:
https://drive.google.com/drive/u/1/folders/15oQXi-hZxMLFGqLtCPVHftqMkciz4DWT (access with berkeley email address)
Important Scripts:
clean_data.py - python script to normalize JSON strings in source data
data_loading.sh - bash script to create folder structure in HDFS and load data from local machine
hive_base_ddl.sql - sql script to create tables and impose schema based on the source data
data_staging.sql - sql script to stage tables for analysis
ML_Model/movie_project_ML_modeling - ipython notebook that constructs Machine Learning Models for collected data