oguzerdo / Level-Based-Persona-Recognition

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Level Based Persona

image-20210516204926891

Business Problem & Purpose

To make new customer definitions by using the characteristics of individual customers (level based persona).

Segmenting new customer definitions.

Integrate system with the new future users.


Project Steps

  • Thinking about the concept of persona.
  • To be able to define new customers according to the levels.
  • Simply segment new customer definitions using the qcut function.
  • When a new customer arrives, classify them according to segments.

Dataset Information

There are two different tables showing the customers' characteristics and transaction information.

users.csv shows the characteristic features of the customers,

purchases.csv shows the transaction information of the customers.

Each user has a unique customer number (uid).

The process of combining both tables can be done with the (uid) number.


Features

users.csv

  • uid: Unique customer number
  • reg_date: Customer's registration date
  • device: Type of product used by the customer. (Android, iOS)
  • gender: Customer’s gender
  • country: Customer's country
  • age: Customer’s age

purchaces.csv

  • uid: Unique customer number
  • date: The date the customer made a purchase
  • price: Customer spend amount

Libraries

pandas
matplotlib

Author

Oğuz Han Erdoğan - oguzerdo


Reference:

VBO - Data Science and Machine Learning Bootcamp
www.veribilimiokulu.com

About


Languages

Language:Jupyter Notebook 96.4%Language:Python 3.6%