Statistics.py

import streamlit as st
import pandas  as pd
import numpy as np
import statistics
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import MinMaxScaler

file = 'KFC.xlsx'
sheet = 'Clean_data'
df = pd.read_excel(file, sheet_name=sheet)

sheet = 'Addon'
df2 = pd.read_excel(file, sheet_name=sheet)

sheet = 'Promotion'
df3 = pd.read_excel(file, sheet_name=sheet)

st.header("Statistical Summary")
st.subheader("Gender Distribution Percentage")
gender_data = df['gender'].value_counts().sort_index()
total = len(df)
mode_value = df['gender'].mode()[0]
col1, col2 = st.columns(2)
with col1:
    st.metric("Total Customers", total)
with col2:
    st.metric("Mode (Most Common)", mode_value)

st.divider()

st.subheader("**Frequency Distribution:**")
for gender, count in gender_data.items():
    percentage = (count / total) * 100
    st.write(f"• **{gender}**: {count} customers ({percentage:.1f}%)")
st.divider()

st.subheader("Nationality Distribution")
nationality_df = df['nationality'].value_counts().sort_index()
mode_value = df['nationality'].mode()[0]
col1, col2 = st.columns(2)
with col1:
    st.metric("Total Count", total)
with col2:
    st.metric("Mode (Most Common)", mode_value)

st.write("---")
st.subheader("Frequency Distribution:")
for nationality_df, count in nationality_df.items():
    percentage = (count / total) * 100
    st.write(f"• **{nationality_df}**: {count} customers ({percentage:.1f}%)")
st.divider()

# DINE TYPE
st.subheader(" Dine Type Distribution")
dinetype_counts = df['dineType'].value_counts().sort_index()
mode_value = df['dineType'].mode()[0]
st.write("Mode:", mode_value)
for dinetype, count in dinetype_counts.items():
    percentage = (count / total) * 100
    st.write(f"• **{dinetype}**: {count} customers ({percentage:.1f}%)")
st.divider()
   
st.subheader("Payment Distribution")
payment_counts = df['payment'].value_counts().sort_index()
mode_value = df['payment'].mode()[0]
st.write("Mode:", mode_value)

for payment, count in payment_counts.items():
    percentage = (count / total) * 100
    st.write(f"• **{payment}**: {count} customers ({percentage:.1f}%)")
    
st.divider()

# Age Statistical Summary 
st.subheader("Statistical Summary of Age")

# Mapping: 1=under 18, 2=18-22, 3=23-27, 4=28-35, 5=above 35

median_age = df['age_enc'].median()
mode_age= df['age_enc'].mode()[0]

st.write('Median (Middle Budget Index):', median_age)
st.write("Mode (Most Frequent Budget Index):", mode_age)
st.write('*Interpretation*')
st.markdown('''
The age data was encoded into five categories, 
where 1 = under 18, 2 = 18–22, 3 = 23–27, 4 = 28–35, and 5 = above 35. 
Based on this, both the median (2.0) and mode (2) fall within the 18–22 age group, 
indicating that this is both the middle and most common age range among respondents. 
This suggests that the majority of participants are young adults, 
with responses largely concentrated in this age group.''')

st.divider()

# Budget Statistical Summary 
st.subheader("Statistical Summary of Spending Budget")

# Mapping: 1=Below 100, 2=100-199, 3=200-299, 4=300+

median_budget = df['budget_enc'].median()
mode_budget = df['budget_enc'].mode()[0]

st.write('Median (Middle Budget Index):', median_budget)
st.write("Mode (Most Frequent Budget Index):", mode_budget)
st.write('*Interpretation*')
st.markdown('''
The budget data was first encoded into four categories, 
where (1 = below 100, 2 = 100–199, 3 = 200–299, and 4 = 300+). 
Based on this, both the median (2.0) and mode (2) fall in the 100–199 range, 
indicating that this is both the middle and most common budget among respondents. 
This suggests that most individuals have a moderate budget, 
with spending behavior largely concentrated within this range.''')

st.divider()

# Visit Frequency Statistical Summary 
st.subheader("Statistical Summary of Visit Frequency")

# Mapping: 1=Rarely, 2=Sometimes, 3=Often, 4=Very Often

median_freq = df['visitFrequency_enc'].median()
mode_freq = df['visitFrequency_enc'].mode()[0]

st.write('Median (Middle Frequency Index):', median_freq)
st.write("Mode (Most Frequent Frequency Index):", mode_freq)

st.write('*Interpretation*')
st.markdown('''
The visit frequency data was encoded into four categories, 
where 1 = Rarely, 2 = Sometimes, 3 = Often, and 4 = Very Often. 
Based on this, both the median (2.0) and mode (2) fall under “Sometimes”, 
indicating that this is both the middle and most common response among participants. 
This suggests that most individuals visit occasionally rather than frequently, 
showing a moderate level of engagement.''')

st.divider()

# Calculate values for flavor and service ratings
mean_flavor = df['flavorRating'].mean()
median_flavor = df['flavorRating'].median()
std_flavor = df['flavorRating'].std()
mode_flavor = df['flavorRating'].mode()[0]

mean_service = df['serviceRating'].mean()
median_service = df['serviceRating'].median()
std_service = df['serviceRating'].std()
mode_service = df['serviceRating'].mode()[0]

# Displaying the results in two columns
col1, col2 = st.columns(2)

with col1:
    st.markdown("**Flavor Rating Statistics**")
    # st.write('Mean:', round(mean_flavor, 2))
    st.write('Median:', median_flavor)
    # st.write('Standard Deviation:', round(std_flavor, 4))
    st.write("Mode:", mode_flavor)

with col2:
    st.markdown("**Service Rating Statistics**")
    # st.write('Mean:', round(mean_service, 2))
    st.write('Median:', median_service)
    # st.write('Standard Deviation:', round(std_service, 4))
    st.write("Mode:", mode_service)

st.divider()