https://github.com/brianruizy/box-office-prediction
Machine Learning - Project using a large historical box office data to predict a movie's 'success' by testing through Regression models.
https://github.com/brianruizy/box-office-prediction
Last synced: 6 months ago
JSON representation
Machine Learning - Project using a large historical box office data to predict a movie's 'success' by testing through Regression models.
- Host: GitHub
- URL: https://github.com/brianruizy/box-office-prediction
- Owner: BrianRuizy
- Created: 2019-03-20T05:58:44.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2019-10-09T05:30:22.000Z (about 6 years ago)
- Last Synced: 2025-02-13T23:20:07.736Z (8 months ago)
- Language: Python
- Homepage:
- Size: 409 KB
- Stars: 0
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Box Office Prediction
## About
This learning project aims at exploring the use of an official Cinema dataset by reading, writing, cleaning and then implementing a Machine Learning model to be trained and predict the success of a movie release. Success can be measured by various different parameters; this program aims at finding relationships between those parameters and consider what makes a movie "succesful". Will then compare the accuracy of both *linear Regression* against *Random Forest Regression*.
Program will undergo this implicit pipeline:
**Cleaning phase** -> **Understanding data** -> **Prediction Models** -> **post-fact data**
## Dependencies
Tools include: [Pandas](https://pandas.pydata.org/),
[Matplotlib](https://matplotlib.org/),
[Numpy](https://www.numpy.org/),
[Seaborn](https://seaborn.pydata.org/),
[SciKit-Learn](https://scikit-learn.org/stable/)
## Data Pipeline
* ### Data Cleansing
At this step we want to remove records that will give us inaccuracies in our results. For instance whenever we have duplicates in the data, or places where we have null & zero values. We proceed by performing the following code...
```python
data = data.drop_duplicates(subset=['move_title', 'title_year'], keep='first').copy()
data.isnull().values.any()
```
* ### Understanding the Data
Here we use matplotlib and pandas to further explore the data set and gain an understanding of the plots. Having done so we can find positive or negative correlations between variables, or find other valubale insights.
* ### ML Models