Skip to content

Latest commit

 

History

History
76 lines (58 loc) · 2.2 KB

README.md

File metadata and controls

76 lines (58 loc) · 2.2 KB
Raw

Information about the python analysis scripts in this directory

Directory structure

data-analysis/
│   README.txt
│   sql_generation.py   
│   data_morphing.py
│   synthetic_data.py
│   knn.py
├── dataset/
├── outputs/
  • sql_generation.py: Responsible for generating the SQL used by the SpringBoot application when creating the in memory DB
  • data_morphing.py: Used to reformat and/or create data to be used by prediction engines
  • synthetic_data.py: Used to create synthetic data to be used by prediction engines
  • knn.py: Current doesn't work that well, in progress
  • dataset/: This directory should be created and populated with the files from the Kaggle dataset found here.
  • outputs/: This directory should be automatically created by scripts that output data.

Dependencies

Scripts written using python 3. Dependencies for the python scripts can be found in requirements.txt

To install the requirements run pip install -r requirements.txt

Tokenized data output (from synthetic_data.py)

Ages

Ages are between 10 and 60, and are generated using a Gaussian distribution with clipping.

The tokens are as follows:

Age range Token
10-19 0
20-29 1
30-39 2
40-49 3
50-60 4

Regions

Regions are generated according to the proportion of Netflix uses from each region.

The tokens are as follows:

Region Token
North America 0
EMEA 1
South America 2
Asia and Oceania 3

Sex

Sexes are generated according to the proportion of Netflix users identifying with each sex.

The tokens are as follows:

Sex Token
Male 0
Female 1
Gender Diverse 2

Relationship status

Relationship statuses were randomly generated.

The tokens are as follows:

Relationship status Token
Married 0
Partner 1
Single 2
Other 3