Project ID: 202212-9
Project Name: Factors Affecting Movie Grossing and Prediction
Group Members: Hanlun Wang (hw2839), Jingtian Zhang (jz3500), Weirui Peng (wp2297)
This repository contains:
- init.py backend file writen in Python based on Flask
- http files frontend files based on Jinja template
- css file define the sytle
- js file define the javascript
- requirements.txt for packages needed
- models models we trained
- crawl.py craws the data on the web page
- combine.py aggregates and pre-processes the collected data
- prediction.ipynb train models and do prediction
This is a system to predict the box office of a movie. It contains data crawling, data visualization, machine learning model training, web backend, and web frontend.
The user can use crawl.py to crawl the data on the web page and combine.py to aggregate and pre-process the collected data. The processed data is stored in the dataset folder.
The user can use predict.py to train and predict the gross. The program will read data from dataset file and automatically split the data into training and predicting sets. Several models are choosen to do the training. All models' training, predicting, and error are shown in the jupyter notebook file.
flaskr file contains the website backend and frontend. Users can see a brief introduction of our project on the website. Users can also input their own data to try the prediction. Their input data will be feeded into the model to do the prediction and the predicted grossing will be shown on the website.
This project uses Python3 and is based on Flask. Use
$ pip install -r requirements.txt
to install required packages.
Go to the flaskr file and run
$ python3 __init__.py