baby names analysis in USA

# Importing pandas and matplotlib and necessary modules
import pandas as pd
import seaborn as sb
import matplotlib.pyplot as plt
import plotly.express as px
import plotly.graph_objects as go
from plotly.subplots import make_subplots
from matplotlib.ticker import EngFormatter
# Read in the baby names CSV as a DataFram
names = pd.read_csv("baby_names.csv")

# Check the data types and missing values
print(names.info())

# Check for unique values in categorical columns
print(names['first_name'].nunique())
print(names['sex'].unique())

# Group by year and sex, then sum the number of babies
yearly_trends_sex = names.groupby(['year', 'sex'])['num'].sum().unstack().reset_index()

# Create an interactive plot
fig = go.Figure()

# Add trace for baby girls
fig.add_trace(go.Scatter(
    x=yearly_trends_sex['year'], 
    y=yearly_trends_sex['F'],
    mode='lines', 
    name='Female', 
    line=dict(color='purple'),
    hovertemplate='Year: %{x}<br>Number of Babies: %{y:,}<extra></extra>'
))

# Add trace for baby boys
fig.add_trace(go.Scatter(
    x=yearly_trends_sex['year'], 
    y=yearly_trends_sex['M'],
    mode='lines', 
    name='Male', 
    line=dict(color='green'),
    hovertemplate='Year: %{x}<br>Number of Babies: %{y:,}<extra></extra>'
))

# Update layout
fig.update_layout(
    title='Total Number of Babies Named Each Year by Gender',
    xaxis_title='Year',
    yaxis_title='Number of Babies',
    yaxis_tickformat=',',  # Use comma as a thousands separator
    legend_title='Gender',
    template='plotly_white'
)

# Show the plot
fig.show()

# Group by 'first_name' and sum the 'num' column
grouped_names = names.groupby('first_name')['num'].sum().reset_index() # this will be use in next cell
grouped_names.rename(columns={'num': 'number'}, inplace=True)
# Order by 'num' in descending order
top_names = grouped_names.sort_values(by='number', ascending=False)
# Display the result
top_names

# Group by 'first_name' and sum the 'num' values
popular_names_sum = names.groupby('first_name')['num'].sum().reset_index()
popular_names_sum.rename(columns={'num': 'number'}, inplace=True)
# Group by 'first_name' and count the number of unique years each name appears in
popular_names_years = names.groupby('first_name')['year'].nunique().reset_index()
# Rename the 'year' column to 'year_count' for clarity
popular_names_years.rename(columns={'year': 'year_count'}, inplace=True)
# Merge the two DataFrames on 'first_name'
popular_names = pd.merge(popular_names_sum, popular_names_years, on='first_name')

# This function can be used to classify names based on their frequency in the dataset.
def classify_popularity(year_count):
    if year_count > 80:
        return 'Classic'
    elif year_count > 50:
        return 'Semi-classic'
    elif year_count > 20:
        return 'Semi-trendy'
    else:
        return 'Trendy'
# Add a new column 'popularity_type' for the classification
popular_names['popularity_type'] = popular_names['year_count'].apply(classify_popularity)
# Drop the 'year_count' column as it's no longer needed
popular_names.drop(columns=['year_count'], inplace=True)
# Order the results alphabetically by 'first_name'
result = popular_names.sort_values(by='first_name')

# Display the result
result.head(20)

# Create a bar chart to visualize the number of names in each popularity type
df = result['popularity_type'].value_counts().reset_index()
df.columns = ['popularity_type', 'count']

fig = px.bar(df, 
             x='popularity_type', 
             y='count', 
             title='Number of Names by Popularity Type',
             labels={'popularity_type': 'Popularity Type', 'count': 'Number of Names'},
             color='popularity_type')  # Add color based on popularity type

# Customize the hover template to show the real number
fig.update_traces(hovertemplate='<b>%{x}</b><br>Number: %{y:,}')

# Show the plot
fig.show()

# Sort the DataFrame by 'number' in descending order and select the top 20 names
top_20 = result.sort_values(by='number', ascending=False).head(20)

# Create an interactive bar chart using Plotly
fig = px.bar(top_20, 
             x='first_name', 
             y='number', 
             color='popularity_type', 
             title='Top 20 Popular Names',
             labels={'first_name': 'First Name', 'number': 'Number of Occurrences'},
             hover_data={'popularity_type': True})

# Customize the hover template to show the real number and popularity type
fig.update_traces(hovertemplate='<b>%{x}</b><br>Number: %{y:,}<br>Popularity: %{customdata[0]}')

# Show the plot
fig.show()

# Merge the classification back to the original DataFrame
names_with_popularity = names.merge(popular_names[['first_name', 'popularity_type']], on='first_name')
# Group by 'year' and 'popularity_type', then count the number of occurrences
popularity_over_years = names_with_popularity.groupby(['year', 'popularity_type']).size().reset_index(name='count')
# Plot the data using Plotly
fig = px.line(popularity_over_years, x='year', y='count', color='popularity_type', 
              title='Popularity Types Over Years', labels={'year': 'Year', 'count': 'total of names'})
# Customize the legend title
fig.update_layout(legend_title_text='Popularity Type')

# Show the plot
fig.show()

# Define a consistent color map for popularity types
color_map = {
    'Classic': '#facc5f' ,  # Yellow
    'Semi-classic': '#43d7a4',  # Green
    'Semi-trendy': '#4095db',  # Blue
    'Trendy': '#6568a0'   # purple
}
# First chart: Number of Names by Popularity Type
df = result['popularity_type'].value_counts().reset_index()
df.columns = ['popularity_type', 'count']
fig1 = px.bar(df,
              x='popularity_type',
              y='count',
              title='Number of Names by Popularity Type',
              labels={'popularity_type': 'Popularity Type', 'count': 'Number of Names'},
              color='popularity_type',
              color_discrete_map=color_map)
fig1.update_traces(hovertemplate='<b>%{x}</b><br>Number: %{y:,}')

# Second chart: Top 20 Popular Names
top_20 = result.sort_values(by='number', ascending=False).head(20)
fig2 = px.bar(top_20,
              x='first_name',
              y='number',
              color='popularity_type',
              title='Top 20 Popular Names',
              labels={'first_name': 'First Name', 'number': 'Number of Occurrences'},
              hover_data={'popularity_type': True},
              color_discrete_map=color_map)
fig2.update_traces(hovertemplate='<b>%{x}</b><br>Number: %{y:,}<br>Popularity: %{customdata[0]}')
# Third chart: Popularity Types Over Years
names_with_popularity = names.merge(popular_names[['first_name', 'popularity_type']], on='first_name')
popularity_over_years = names_with_popularity.groupby(['year', 'popularity_type']).size().reset_index(name='count')
fig3 = px.line(popularity_over_years,
               x='year',
               y='count',
               color='popularity_type',
               title='Popularity Types Over Years',
               labels={'year': 'Year', 'count': 'Total of Names'},
               color_discrete_map=color_map)
fig3.update_layout(legend_title_text='Popularity Type')

# Create subplots
fig = make_subplots(rows=2, cols=2, subplot_titles=("Number of Names by Popularity Type", "Top 20 Popular Names", "Popularity Types Over Years"))

# Add the first chart to the first subplot
for trace in fig1['data']:
    fig.add_trace(trace, row=1, col=1)

# Add the second chart to the second subplot
for trace in fig2['data']:
    fig.add_trace(trace, row=1, col=2)

# Add the third chart to the third subplot
for trace in fig3['data']:
    fig.add_trace(trace, row=2, col=1)

# Update layout
fig.update_layout(title_text="Comparison of Popularity Types and Top 20 Names", showlegend=True, height=800)

# Show the plot
fig.show()

# Assuming 'all_years' is a set of all possible years in the dataset
all_years = set(names['year'].unique())

# Function to find missing years for each name
def find_missing_years(name):
    name_years = set(names[names['first_name'] == name]['year'].unique())
    missing_years = all_years - name_years
    return sorted(missing_years)

# Function to find the first year each name appears
def find_first_year(name):
    name_years = names[names['first_name'] == name]['year']
    if not name_years.empty:
        return name_years.min()
    return None

# Create a DataFrame to store the missing years and first year for each name by popularity type
results = []
for popularity_type in popular_names['popularity_type'].unique():
    type_names = popular_names[popular_names['popularity_type'] == popularity_type]
    missing_years_df = pd.DataFrame({
        'first_name': type_names['first_name'],
        'popularity_type': popularity_type,
        'missing_years': type_names['first_name'].apply(find_missing_years),
        'first_year': type_names['first_name'].apply(find_first_year)
    })
    missing_years_df['missing_count'] = missing_years_df['missing_years'].apply(len)
    results.append(missing_years_df)

# Concatenate all results into a single DataFrame
all_missing_years_df = pd.concat(results)

# Display the full results for all popularity types
all_missing_years_df[['first_name', 'popularity_type', 'first_year', 'missing_count']].sort_values(by=['popularity_type', 'missing_count'], ascending=[True, False])

# Filter the names dataframe to include only female names
female_names = names[names['sex'] == 'F']

# Group by first_name and sum the occurrences
female_name_counts = female_names.groupby('first_name')['num'].sum().reset_index()

# Sort the names by the number of occurrences in descending order
top_female_names = female_name_counts.sort_values(by='num', ascending=False)

# Display the top-ranked female names
top_female_names.head(10)