The Following is a general presentaion of our project
Explore the docs »
View Demo
·
Report Bug
·
Request Feature
-
Shopium Customer Behavior Prediction is a machine learning API made for Shopium to better understand our customers orientation.
-
This module is being developped based on a logistic regression model using Python to study our clients likes and views on offers in a relation with their ages and sex.
-
This project was designed to run on Shopium's fake database generated by Shopium Faker
This section should list any major frameworks/libraries used to bootstrap your project. Leave any add-ons/plugins for the acknowledgements section. Here are a few examples.
-
This project was designed to predict customers behavior based on their ages and sex in a relation with their liked products and offers views.
-
Logistic Regression model was trained using sklearn and Shopium's fake data to test this whole module.
-
This module will take an array of offers IDs and return an array of the recommended offers IDs for a given client based on his/her age and/or sex.
### Prerequisites
This is a list of different main modules to install before implementing our project
- pip
npm install npm@latest -g
- Tesseract
npm install npm@latest -g
Below is an example of how you can instruct your audience on installing and setting up your app. This template doesn't rely on any external dependencies or services.
-
Clone the repo
git clone https://github.com/firas122/CustBehave
-
Install pip packages
pip install -r requirements.txt
-
Run the API using command above (the application will be running on localhost ip address using 5000 as port):
python /project_directory_path/CustBehave/api.py
-
Terminal output will include an url by default 127.0.0.1 (localhost) running on port 5000 using the path /predict
-
Send a
POST
request to that url with three variablesDB_url
which contains the url to your mongo database,user_id
that represents the userwe want to make recommendations for andoffers_array
an array of available offers for recommendations . -
The returned result should be an array of offers ids to recommend as the example above :
- And the returning JSON Object should include possible matchs with offers :
Shopium - @Shopium - shopium.local@gmail.com
Project Link: https://github.com/firas122/Scan