import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import warnings
warnings.filterwarnings('ignore')

# Example: Load the Social Media Content dataset
np.random.seed(42)
n_posts = 500
views = np.random.randint(100, 100000, n_posts)
df = pd.DataFrame({
    'post_id': range(1, n_posts+1),
    'platform': np.random.choice(['YouTube','Instagram','TikTok'], n_posts),
    'date': pd.date_range('2023-01-01', periods=n_posts, freq='6h'),
    'views': views,
    'likes': (views * np.random.uniform(0.02, 0.15, n_posts)).astype(int),
    'comments': (views * np.random.uniform(0.001, 0.05, n_posts)).astype(int),
    'shares': (views * np.random.uniform(0.001, 0.03, n_posts)).astype(int)
})
print(df.shape)
df.head(3)

# Calculate engagement rate for each post
df['engagement_rate'] = (df['likes'] + df['comments'] + df['shares']) / df['views']
print('Mean engagement rate: {:.2%}'.format(df['engagement_rate'].mean()))
df[['platform','engagement_rate']].head()

# Group by platform to compare engagement rates
platform_avg_engagement = df.groupby('platform')['engagement_rate'].mean().sort_values(ascending=False)
print(platform_avg_engagement)
platform_avg_engagement.plot(kind='bar', color=['#FF0000','#C13584','#69C9D0'])
plt.ylabel('Average Engagement Rate')
plt.title('Average Engagement Rate by Platform')
plt.show()

# Calculate daily average engagement rate
df['date_only'] = df['date'].dt.date
daily_trend = df.groupby('date_only')['engagement_rate'].mean()
plt.figure(figsize=(10,4))
daily_trend.plot()
plt.ylabel('Avg Engagement Rate')
plt.xlabel('Date')
plt.title('Daily Engagement Rate Trend')
plt.tight_layout()
plt.show()

# Get top 10 posts by engagement rate
top_posts = df.sort_values('engagement_rate', ascending=False).head(10)
print(top_posts[['platform','date','views','engagement_rate']])

# Plot engagement trends by platform
platform_trend = df.groupby(['date_only','platform'])['engagement_rate'].mean().unstack()
platform_trend.plot(figsize=(12,6))
plt.ylabel('Avg Engagement Rate')
plt.title('Engagement Rate Trend by Platform')
plt.legend(title='Platform')
plt.tight_layout()
plt.show()

# Boxplot for engagement rate across platforms
plt.figure(figsize=(8,5))
sns.boxplot(x='platform', y='engagement_rate', data=df, palette='pastel')
plt.ylabel('Engagement Rate')
plt.title('Engagement Rate Distribution per Platform')
plt.show()

# Extract hour from date and analyze average engagement rate by posting hour
df['post_hour'] = df['date'].dt.hour
hourly_engagement = df.groupby('post_hour')['engagement_rate'].mean()
plt.figure(figsize=(8,4))
hourly_engagement.plot(kind='bar', color='#36a2eb')
plt.xlabel('Hour of Day')
plt.ylabel('Mean Engagement Rate')
plt.title('Average Engagement Rate by Posting Hour')
plt.tight_layout()
plt.show()

# Detect viral posts: posts above the 99th percentile engagement rate
threshold = df['engagement_rate'].quantile(0.99)
viral_posts = df[df['engagement_rate'] >= threshold]
print('Number of viral posts:', len(viral_posts))
viral_posts[['platform','date','views','engagement_rate']]

# Introduce missing values artificially for demonstration
df_missing = df.copy()
df_missing.loc[np.random.choice(df_missing.index, size=10), 'likes'] = np.nan
print('Missing likes before:', df_missing["likes"].isna().sum())
df_missing['likes'] = df_missing['likes'].fillna(0)
print('Missing likes after:', df_missing["likes"].isna().sum())

# Incorrect: Summing engagement rates
agg_sum = df.groupby('platform')['engagement_rate'].sum()
print('Summing engagement rates (incorrect):\n', agg_sum)
# Correct: Taking mean engagement rate
agg_mean = df.groupby('platform')['engagement_rate'].mean()
print('\nMean engagement rates (correct):\n', agg_mean)

# Monthly engagement rate benchmarking
df['month'] = df['date'].dt.to_period('M')
monthly_mean = df.groupby('month')['engagement_rate'].mean()
print(monthly_mean.tail(2))
monthly_mean.plot(marker='o', linestyle='-', figsize=(10,4))
plt.title('Monthly Engagement Rate Trend')
plt.ylabel('Average Engagement Rate')
plt.xlabel('Month')
plt.tight_layout()
plt.show()

# Define a function for engagement rate
def calc_engagement_rate(df):
    return (df['likes'] + df['comments'] + df['shares']) / df['views']
df['engagement_rate_2'] = calc_engagement_rate(df)
print('Results equal:', np.allclose(df['engagement_rate'], df['engagement_rate_2']))

# Filter posts from last 30 days and sort by engagement rate
latest_date = df['date'].max().date()
window_start = latest_date - pd.Timedelta(days=30)
recent = df[df['date'].dt.date >= window_start]
top_recent = recent.sort_values('engagement_rate', ascending=False).head(5)
print(top_recent[['platform','date','views','engagement_rate']])

plt.figure(figsize=(8, 4))
plt.bar(top_recent['date'].dt.strftime('%Y-%m-%d'), top_recent['engagement_rate'], color='#2ecc71')
plt.xticks(rotation=45)
plt.ylabel('Engagement Rate')
plt.title('Top 5 Recent Posts: Engagement Rate')
plt.tight_layout()
plt.show()