Spaces:
Sleeping
Sleeping
File size: 6,915 Bytes
e7e98f2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 |
import streamlit as st
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import plotly.express as px
from PIL import Image
st.set_page_config(
page_title='Churn Predictor',
layout='wide',
initial_sidebar_state='expanded'
)
def run():
# judul
st.title('**Churn Exploration**')
st.subheader('Explore The Churn Dataset')
# tambah gambar
image = Image.open('churn.jpg')
st.image(image)
st.markdown('---')
markdown_text = '''
## Background
In today's competitive business landscape, customer churn has become a
significant concern for many companies. Customer churn refers to the
phenomenon where customers discontinue using a company's products or
services. Churn can have a negative impact on a company's revenue,
growth, and overall success. Therefore, companies are increasingly
focused on identifying customers who are likely to churn so that they
can take proactive measures to retain them.
## Objective
The objective of this project is to develop a deep learning model for
churn prediction. The company wants to minimize the risk of customer
churn by accurately predicting which customers are likely to stop using
their products or services. By identifying potential churners in advance,
the company can take targeted actions and implement retention strategies
to reduce churn rates and maximize customer loyalty.
## About Dataset
| Variable | Description |
|-------------------------|-----------------------------------------------------------------------------------------------|
| user_id | ID of a customer |
| age | Age of a customer |
| gender | Gender of a customer |
| region category | Region that a customer belongs to |
| membership category | Category of the membership that a customer is using |
| joining date | Date when a customer became a member |
| joined through referal | Whether a customer joined using any referral code or ID |
| preferred_offer types | Type of offer that a customer prefers |
| medium_of operation | Medium of operation that a customer uses for transactions |
| internet option | Type of internet service a customer uses |
| last visit time | The last time a customer visited the website |
| days since last login | Number of days since a customer last logged into the website |
| average time spent | Average time spent by a customer on the website |
| average transaction | Average transaction value of a customer |
| average freq login days | Number of times a customer has logged in to the website |
| point in wallet | Points awarded to a customer on each transaction |
| used spesial discount | Whether a customer uses special discounts offered |
| offer app preference | Whether a customer prefers offers |
| past complaint | Whether a customer has raised any complaints |
| complaint status | Whether the complaints raised by a customer was resolved |
| feedback | Feedback provided by a customer |
| churn risk score | Churn Score |
'''
st.markdown(markdown_text)
st.markdown('---')
st.subheader('**Data Exploratory**')
st.markdown('---')
st.write('### Customer Information')
# show dataset
data = pd.read_csv('churn.csv')
st.dataframe(data)
st.markdown('---')
st.write("### Today's Condition")
st.markdown('---')
# show distribusi customer churn
fig, ax = plt.subplots()
plt.pie(data['churn_risk_score'].value_counts(),
labels=['Churn', 'Not-Churn'],
autopct='%1.1f%%',
colors=['Grey', 'Orange'],
startangle=40,
explode=[0.05, 0])
plt.title('Customer Churn Percentage')
plt.axis('equal')
st.pyplot(fig)
'''
Based on the above graph, it can be observed that the distribution of Churn Risk
Score ***tends to be evenly divided*** among its values, indicating that the feature
follows a normal distribution
'''
# visual numerical
st.subheader('Chart Based on Metrics')
st.markdown('---')
choice = st.selectbox('Pick Numeric Columns: ', ('age', 'days_since_last_login', 'avg_time_spent',
'avg_transaction_value', 'avg_frequency_login_days', 'points_in_wallet'))
fig, ax = plt.subplots(figsize=(8, 6))
sns.kdeplot(data=data, x=choice, fill=True,
hue='churn_risk_score', palette='inferno')
ax.set_title(choice.capitalize()+' Ratio')
st.pyplot(fig)
st.markdown('---')
# visual categorical
choice_2 = st.selectbox('Pick Category Column : ', ('gender', 'region_category',
'membership_category', 'joined_through_referral', 'preferred_offer_types',
'medium_of_operation', 'internet_option', 'used_special_discount',
'offer_application_preference', 'past_complaint', 'complaint_status', 'feedback'))
fig = plt.figure(figsize=(15, 10))
sns.countplot(data=data, x=choice_2,
hue='churn_risk_score', palette='viridis')
plt.xlabel(choice_2.capitalize())
plt.ylabel('Count')
plt.title(choice_2.capitalize()+' Ratio')
plt.legend(title='Churn Risk Score')
st.pyplot(fig)
if __name__ == '__main__':
run()
|