Skip to content

Customer Segmentation Recommendation is a Machine Learning based project which analysis user data and preferences to provide product recommendations on a micro-segment level and also helps the marketing team in making better marketing-business strategy by analysing its customers based on their buying behaviour.

License

Notifications You must be signed in to change notification settings

bhaveshsingh0206/customer-segmentation-recommendation

Repository files navigation

Customer Segmentation Recommendation

About

Customer Segmentation Recommendation is a Machine Learning based project which analyzes user data and preferences to provide product recommendations on a micro-segment level and also helps the marketing team in making better marketing-business strategy by analysing its customers, based on their buying behaviour.

Micro segmentation is the process of dividing a market of potential customers into groups, or segments, based on different characteristics.

Types of Market Segmentation:

  • Demographic segmentation( age, gender, income, etc.)
  • Psychographic segmentation( interests, lifestyle, etc.)
  • Behavioral segmentation( purchase & spend habits, brand interactions, etc.)
  • Geographic segmentation( city, country, etc.)

Based on data and after analysing, users were divided into various categories such as

  • Tech (Tech Savy, Luxuorios, Flagship Killers, Others..)
  • Brand Lovers (Apple, Samsung, Huawei, Others..)
  • Camera Lovers (Sensible, PhotoPhile, SelfieLover)

Technical Architecture

Architecture

Technology Stack

  • Web Frontend developed using Html/css and Vanilla JavaScript

  • Backend developed using Flask

  • Machine Learning Classfication and Regression Algorithms, Neural Networks

  • Models trained by scrapping data from the Internet

  • Dataset comprises of users previous purchase history, product specifications, and user data on a micro-segment.

    • Data stored in excel

Features

Marketing Head

  • Analyse Buying Behavior of customers based on various paramters and divide them into various categories.
  • Make better marketing strategy based on customers like/dislike for a particular product.
  • Analyse a product sale and take actions to increase it.
  • Predict the performance of their product based on its specification(in this case mobile phones).
  • Target a specific audience based on product performance.
  • Predict the audience that should be targeted on the given product.
  • Recommend similar products to customers.

Screenshots

Dashboard (Analytics)

1 2 3 4 5

Prediction (Performance and User Category)

pred1 pred2 pred3

Recommendations

recom recom

Configure Project

  • Install python on your machine python ver 3.5.9 link to download : python.org/downloads/

  • After successful installation of python, install following python libraries in your machine

     Package               Version
     --------------------- --------
    
     matplotlib            3.1.2
     missingno             0.4.2
     numpy                 1.18.1
     pandas                1.0.0
     sklearn               0.0
     tensorflow 2.0        1.15
     keras                 2.3.0
     seaborn               0.10.0
     flask                 1.1.1
     flask-Cors            3.0.8
    
     How to install?
     Open command prompt as administrator and type command pip install 'library-name'(For Windows)
    
  • After successful installation of following libraries,navigate to the "DBS-Segmentation" folder where your project is kept,Open the folder and run "server.py" file Running this file will start the project server

  • After successfully running "server.py" file, go to "frontend folder ->html folder ->dbshome.html",run "dbshome.html" file,thus the project will be setup successfully

  • Inorder to have better understanding of code,install Anaconda-Navigator after that open jupyter-notebooks and open this folder "python-notebooks" on jupyter notebook and run files accordingly in that folder

About

Customer Segmentation Recommendation is a Machine Learning based project which analysis user data and preferences to provide product recommendations on a micro-segment level and also helps the marketing team in making better marketing-business strategy by analysing its customers based on their buying behaviour.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published