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

np.random.seed(42)
df = pd.DataFrame({'CustomerID': range(1,501), 'Age': np.random.randint(18,70,500), 'Region': np.random.choice(['North','South','East','West'],500), 'NPS_Score': np.random.randint(0,11,500)})
print(df.shape)
print(df.head(3))

plt.figure(figsize=(7,4))
sns.histplot(df['NPS_Score'], bins=11, kde=False, color='skyblue')
plt.title('Distribution of NPS Scores')
plt.xlabel('NPS Score (0-10)')
plt.ylabel('Number of Customers')
plt.tight_layout()
plt.show()

# Classify NPS responses
def nps_category(score):
    if score >= 9:
        return 'Promoter'
    elif score >= 7:
        return 'Passive'
    else:
        return 'Detractor'
df['NPS_Type'] = df['NPS_Score'].apply(nps_category)
counts = df['NPS_Type'].value_counts()
print(counts)

n_promoters = counts.get('Promoter', 0)
n_detractors = counts.get('Detractor', 0)
n_total = len(df)
nps_score = ((n_promoters - n_detractors) / n_total) * 100
print(f"The company's Net Promoter Score (NPS) is: {nps_score:.2f}")

plt.figure(figsize=(8,4))
sns.countplot(data=df, x='Region', hue='NPS_Type', palette='coolwarm')
plt.title('NPS Segments by Region')
plt.xlabel('Region')
plt.ylabel('Customer Count')
plt.tight_layout()
plt.show()

df_feedback = pd.DataFrame({'CustomerID':[1,2,3,4,5], 'Feedback':['Great service and friendly staff','Delivery was slow and packaging was poor','Excellent quality, will buy again','Customer support needs improvement','Good value for money']})
print(df_feedback.head(3))

keywords = {}
for text in df_feedback['Feedback']:
    for word in text.lower().split():
        keywords[word] = keywords.get(word, 0) + 1
sorted_keywords = sorted(keywords.items(), key=lambda x: x[1], reverse=True)
print('Most common words:')
for word, count in sorted_keywords[:5]:
    print(f'{word}: {count}')

dataset = openml.datasets.get_dataset(1461)
df_marketing, _, _, _ = dataset.get_data(dataset_format='dataframe')
df_marketing.columns = ['age','job','marital','education','default','balance','housing','loan','contact','day','month','duration','campaign','pdays','previous','poutcome','response']
print(df_marketing.shape)
print(df_marketing.head(3))

response_rate = df_marketing['response'].value_counts(normalize=True) * 100
print('Campaign Response Rate (%):')
print(response_rate.round(2))

plt.figure(figsize=(8,4))
sns.barplot(x='education', y='response', data=df_marketing.replace({'response': {'yes': 1, 'no': 0}}), estimator=np.mean)
plt.title('Campaign Response Rate by Education Level')
plt.xlabel('Education Level')
plt.ylabel('Response Rate')
plt.tight_layout()
plt.show()

merged = pd.merge(df, df_marketing[['response']].iloc[:len(df)], left_index=True, right_index=True)
xtab = pd.crosstab(merged['NPS_Type'], merged['response'], normalize='index') * 100
print('Cross-tabulation: NPS Segment vs. Campaign Response')
print(xtab.round(1))

df['Age_Group'] = pd.cut(df['Age'], bins=[17,30,45,60,70], labels=['18-30','31-45','46-60','61-70'])
plt.figure(figsize=(7,4))
sns.boxplot(x='Age_Group', y='NPS_Score', data=df, palette='pastel')
plt.title('Distribution of NPS Scores by Age Group')
plt.xlabel('Age Group')
plt.ylabel('NPS Score')
plt.tight_layout()
plt.show()

positive_words = ['great', 'excellent', 'good', 'friendly']
negative_words = ['poor', 'slow', 'needs improvement', 'bad']
def simple_sentiment(text):
    text = text.lower()
    if any(w in text for w in positive_words):
        return 'Positive'
    elif any(w in text for w in negative_words):
        return 'Negative'
    else:
        return 'Neutral'
df_feedback['Sentiment'] = df_feedback['Feedback'].apply(simple_sentiment)
print(df_feedback[['Feedback','Sentiment']])

np.random.seed(42)
dates = pd.date_range('2022-01-01', periods=12, freq='M')
trend_data = pd.DataFrame({'Month': dates})
trend_data['Mean_NPS'] = np.random.normal(7, 1, 12).clip(0,10)
plt.figure(figsize=(8,4))
plt.plot(trend_data['Month'], trend_data['Mean_NPS'], marker='o', color='navy')
plt.title('NPS Trend Over Time')
plt.xlabel('Month')
plt.ylabel('Mean NPS Score')
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

import openml
dataset = openml.datasets.get_dataset(42178)
df_satisfaction, _, _, _ = dataset.get_data(dataset_format='dataframe')
factors = ['OnlineSecurity', 'TechSupport', 'StreamingTV', 'OnlineBackup', 'Churn']
for col in factors:
    df_satisfaction[col] = df_satisfaction[col].astype(str)
    df_satisfaction[col] = df_satisfaction[col].map({'Yes': 1, 'No': 0, 'No internet service': 0, 'No phone service': 0, 'nan': 0, 'False': 0, 'True':1}).fillna(0)
df_satisfaction['Brand_Perception_Index'] = df_satisfaction[factors[:-1]].mean(axis=1) * (1 - df_satisfaction['Churn'])
print(df_satisfaction[['Brand_Perception_Index']].head(5))
print('Index ranges from 0 (poor perception) to 1 (very positive perception and retention)')

with open('brand_perception_report.txt', 'w') as f:
    f.write('Brand Perception Analysis Report\n')
    f.write(f'NPS: {nps_score:.2f}\n')
    f.write(f'Highest NPS Region: {df.groupby("Region")["NPS_Score"].mean().idxmax()}\n')
    f.write('Sample Brand Perception Indices (first 5):\n')
    for idx, val in enumerate(df_satisfaction["Brand_Perception_Index"].head(5)):
        f.write(f'Customer {idx+1}: {val:.2f}\n')

df_missing = df.copy()
df_missing.loc[5:10, 'NPS_Score'] = np.nan
missing_count = df_missing['NPS_Score'].isnull().sum()
print(f'Missing NPS responses: {missing_count}')
df_missing['NPS_Score'] = df_missing['NPS_Score'].fillna(df['NPS_Score'].mean())
print('Filled missing scores with mean NPS.')

def mistake_nps_group(score):
    if score > 8:
        return 'Promoter'
    elif 6 < score <= 8:
        return 'Passive'
    else:
        return 'Detractor'
df['Wrong_NPS_Type'] = df['NPS_Score'].apply(mistake_nps_group)
error_count = (df['NPS_Type'] != df['Wrong_NPS_Type']).sum()
print(f'Number of records affected by grouping error: {error_count}')

region_sum = df.groupby('Region')['NPS_Score'].sum()
region_mean = df.groupby('Region')['NPS_Score'].mean()
print('Total NPS by Region (should use mean for insight):')
print(region_sum)
print('Correct: Mean NPS by Region:')
print(region_mean)

crosstab = pd.crosstab(df['Age_Group'], df['NPS_Type'], normalize='index') * 100
print('NPS Distribution by Age Group (%)')
print(crosstab.round(1))

index_mean = df_satisfaction['Brand_Perception_Index'].mean()
index_trend = df_satisfaction.groupby('SeniorCitizen')['Brand_Perception_Index'].mean()
print(f'Mean Brand Perception Index: {index_mean:.2f}')
print('Index by SeniorCitizen:')
print(index_trend.round(2))

# Calculate mean NPS and highest sentiment from all inputs
overall_nps = df['NPS_Score'].mean()
sentiment_counts = df_feedback['Sentiment'].value_counts()
if sentiment_counts.get('Negative',0) > sentiment_counts.get('Positive',0):
    rec = 'Address complaints in support and delivery for brand improvement.'
else:
    rec = 'Maintain current service strengths while promoting positive attributes.'
print(f'Average NPS: {overall_nps:.2f}')
print(f'Feedback Sentiment Breakdown: {dict(sentiment_counts)}')
print(f'Recommendation: {rec}')