Skip to content

AshNumpy/DL-Churn-Analysis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

Click for Turkish Description

Projeye Genel Bakış:

Bu proje, bir bankanın müşteri churn (müşteri kaybı) durumunu analiz etmeyi ve tahmin etmeyi amaçlamaktadır. Churn, müşterilerin belirli bir süre içinde bankayı terk etmesi anlamına gelir. Müşteri churn oranlarının belirlenmesi ve müşteri kayıplarının önceden tahmin edilmesi, bankalar için stratejik öneme sahiptir. Bu proje, müşteri davranışlarını inceleyerek, müşterilerin bankayı terk etme olasılığını tahmin eden bir makine öğrenmesi modeli geliştirmeyi amaçlamaktadır.

Projenin Hedefleri:

  • Müşteri Churn Analizi: Müşteri churn durumunu etkileyen faktörler ve müşteri davranışlarını belirlemek için veri analizi yapmak.

  • Tahmin Modeli Oluşturma: Keras kütüphanesi kullanılarak bir makine öğrenmesi modeli geliştirmek ve müşterilerin churn etme olasılığını tahmin etmek.

  • Performans Değerlendirmesi: Oluşturulan tahmin modelinin performansını değerlendirmek ve doğruluk, hassasiyet ve geri çağırma gibi metriklerle değerlendirmek.

Projenin Özellikleri:

  • Veri Toplama ve Önişleme: Müşteri verilerini toplamak ve ön işleme adımlarıyla eksik verileri ele almak, veriyi temizlemek ve modele uygun hale getirmek.

  • Makine Öğrenmesi Modeli: Keras kütüphanesindeki yapay sinir ağı algoritmalarını kullanarak bir sınıflandırma modeli oluşturmak.

  • Performans Değerlendirmesi: Oluşturulan modelin performansını k-fold cross-validation ve farklı metriklerle değerlendirmek.

  • Model Optimizasyonu: Hyperparameter tuning ve feature selection teknikleri kullanarak modelin performansını artırmak.

  • Sonuçların Görselleştirilmesi: Proje sonuçlarını grafikler ve görsellerle raporlamak.

Project Overview:

This project aims to analyze and predict customer churn for a bank. Churn refers to customers leaving the bank within a certain period of time. Determining customer churn rates and predicting customer losses in advance are of strategic importance for banks. The project aims to develop a machine learning model that predicts the likelihood of customers churning by examining customer behaviors.

Project Objectives:

  • Customer Churn Analysis: Analyzing factors and customer behaviors that influence churn to understand customer attrition patterns.

  • Building the Prediction Model: Developing a machine learning model using the Keras library to predict the probability of customers churning.

  • Performance Evaluation: Evaluating the performance of the prediction model using metrics such as accuracy, precision, and recall.

Project Features:

  • Data Collection and Preprocessing: Collecting customer data and preprocessing it by handling missing data, cleaning the data, and preparing it for modeling.

  • Machine Learning Model: Building a classification model using artificial neural network algorithms available in the Keras library.

  • Performance Evaluation: Evaluating the model's performance using k-fold cross-validation and various metrics.

  • Model Optimization: Improving the model's performance through hyperparameter tuning and feature selection techniques.

  • Visualization of Results: Reporting project outcomes using graphs and visualizations.

Feel free to review and make any necessary adjustments to the provided answers. If you need any further assistance, feel free to ask.