IE6400_Quiz21_Day23

Product-Specific Customer Sentiment Analysis Data Preparation: Load and examine the dataset, focusing on the structure relevant to each product. import pandas as pd import random import faker def generate_customer_comments ( num_comments = 500 ): # Initialize Faker for generating fake data fake = faker . Faker () # Define sample product names and IDs product_names = [ 'AlphaPhone' , 'BetaBook Pro' , 'GammaPad' , 'Delta Earbuds' , 'Epsilon Charger' ] product_ids = range ( 1001 , 1006 ) # Sample comments comments = [ "Great product, very satisfied!" , "Could be better." , "Not what I expected." , "Fantastic quality, will buy again!" , "Poor customer service." , "Excellent value for money." , "Disappointed with the purchase." , "Exceeded my expectations." , "Shipping was slow." , "Absolutely love this product!" ] # Generate the dataset data = [] for _ in range ( num_comments ): customer_id = fake . random_number ( digits = 5 ) first_name = fake . first_name () last_name = fake . last_name () product_name = random . choice ( product_names ) product_id = random . choice ( product_ids ) comment = random . choice ( comments ) data . append ([ customer_id , first_name , last_name , product_name , product_id , comment ]) # Create DataFrame df = pd . DataFrame ( data , columns = [ 'CustomerID' , 'FirstName' , 'LastName' , 'ProductName' , 'ProductID' , 'Comment' ]) return df # Generate the dataset df_comments = generate_customer_comments () df_comments . head () CustomerID FirstName LastName ProductName ProductID Comment 0 17767 Megan Brooks Delta Earbuds 1002 Disappointed with the purchase. 1 71559 Christopher Scott BetaBook Pro 1002 Absolutely love this product ! 2 56682 Stephen Conner GammaPad 1001 Absolutely love this product ! 3 55352 Leslie Smith AlphaPhone 1002 Excellent value for money. 4 60923 Katrina King Epsilon Charger 1003 Exceeded my expectations. Sentiment Classification: Develop a model to classify comments into sentiments (positive, neutral, negative) for each product. from textblob import TextBlob def get_sentiment ( text ): analysis = TextBlob ( text ) if analysis . sentiment . polarity > 0 : return 'positive' elif analysis . sentiment . polarity == 0 : return 'neutral' else : return 'negative' # Apply sentiment classification to the 'Comment' column df_comments [ 'Sentiment' ] = df_comments [ 'Comment' ] . apply ( get_sentiment ) import matplotlib.pyplot as plt import seaborn as sns # Create a bar plot for sentiment distribution by product plt . figure ( figsize = ( 12 , 6 )) ax = sns . countplot ( x = 'ProductName' , hue = 'Sentiment' , data = df_comments ) plt . title ( 'Sentiment Distribution by Product' ) # Add count labels to each bar for p in ax . patches : ax . annotate ( f'{ p . get_height () }' , ( p . get_x () + p . get_width () / 2. , p . get_height ()), ha = 'center' , va = 'center' , xytext = ( 0 , 10 ), textcoords = 'offset points' ) plt . show () Analysis and Insights: Interpret the results, compare sentiments across different products. # Print Each product polarity count product_sentiment_counts = df_comments . groupby ([ 'ProductName' , 'Sentiment' ]) . size () . unstack ( fill_value = 0 ) print ( product_sentiment_counts ) Sentiment negative neutral positive ProductName AlphaPhone 45 9 45 BetaBook Pro 34 8 41 Delta Earbuds 52 13 43 Epsilon Charger 34 11 48 GammaPad 46 11 60 Interpretation: GammaPad has the highest count of positive sentiments, followed by Epsilon Charger and AlphaPhone. Delta Earbuds have a higher count of negative sentiments compared to other products. AlphaPhone shows a balanced distribution between negative and positive sentiments. Visualization: Create visual representations to illustrate sentiment distribution for each product. # Function to get sentiment score def get_sentiment_score ( comment ): analysis = TextBlob ( comment ) return analysis . sentiment . polarity # Apply sentiment scoring to the 'Comment' column df_comments [ 'SentimentScore' ] = df_comments [ 'Comment' ] . apply ( get_sentiment_score ) avg_sentiment_per_product = df_comments . groupby ( 'ProductName' )[ 'SentimentScore' ] . mean () . reset_index () # Create a bar plot for average sentiment score per product plt . figure ( figsize = ( 10 , 6 )) sns . barplot ( x = 'ProductName' , y = 'SentimentScore' , data = avg_sentiment_per_product ) plt . title ( 'Average Sentiment Score per Product' ) plt . ylabel ( 'Average Sentiment Score' ) plt . xlabel ( 'Product Name' ) plt . show () In [1]: Out[1]: In [2]: In [4]: In [5]: In [6]: In [7]: In [8]: Average Sentiment Score per Product 0.25 - 0.20 - B 。 1 0 - Avel 0.05 - 0.00 AlphaPhone BetaBook Pro Delta Earbuds Product Name Epsilon Charger GammaPad In [ ]: