Sentiment Analysis Case Study

  The Background CouchPotato Couriers, a unique delivery service known for its relaxed and humorous approach to deliveries, places a high priority on ensuring that customers can truly "sit back and relax." As they continue to grow, they recognize the importance of understanding customer sentiments to enhance their service and maintain a lighthearted customer experience. To maintain their unique and fun approach while ensuring service excellence, they need to gauge customer sentiments accurately and act on feedback to address concerns. You are an experienced data analyst working for CouchPotato Couriers, so you enthusiastically volunteer to analyze customer feedback data. Unfortunately, the data is not labelled, but since you are an experienced data analyst, that should not be a problem. The Data Along with this write-up, you should have received one dataset: 1. cc_customer_feedback.csv This dataset contains feedback from 674 customers from the month of October 2023. This dataset contains two columns, one column containing the feedback text and the other containing the date. The Case Study Your manager has given you the following task: 1. Perform exploratory analysis to provide insights into the data. 2. Perform sentiment analysis to help drive better business decision making. In particular, your manager wants to know: a. What is the overall sentiment for our organization? b. How has the sentiment changed over time? c. Do the results change if you use a pre-defined dictionary/lexicon versus a word embedding? (Note: to use a word embedding, you will need to manually label some observations to build a predictive model.). Note: Undergraduate students only need to use a dictionary/lexicon. Graduate students must use a dictionary/lexicon and a word embedding. 3. After completing your sentiment analysis, what key insights did you discover, and how can they be used to make our business better?  

import pandas as pd

# Load the data into a DataFrame
df = pd.read_csv('cc_customer_feedback.csv')

df.head()

                                feedback_text                date
0                      "Thank you for the prompt delivery!" 2023-10-01
1                                         "Great service!"   2023-10-01
2    "The delivery guy was so friendly and helpful. Thanks!" 2023-10-01
3  "I'm always satisfied with CouchPotato Couriers. :)"    2023-10-01
4                       "My order arrived late and cold."   2023-10-01

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 674 entries, 0 to 673
Data columns (total 2 columns):
 #   Column       Non-null Count    Dtype 
---  ------       --------------    -----
 0  feedback_text    674 non-null    object
 1  date          674 non-null    object
dtypes: object(2)
memory usage: 107.2 KB

import nltk
from nltk.corpus import stopwords
from nltk.stem import PorterStemmer

# Remove punctuation
df['feedback_text'] = df['feedback_text'].apply(lambda text: re.sub(r'[^\w\s]', '', text))

# Convert text to lowercase
df['feedback_text'] = df['feedback_text'].apply(lambda text: text.lower())

# Remove stop words
stop_words = set(stopwords.words('english'))
df['feedback_text'] = df['feedback_text'].apply(lambda text: [word for word in text.split() if word not in stop_words])

# Stemming
stemmer = PorterStemmer()
df['feedback_text'] = df['feedback_text'].apply(lambda text: [stemmer.stem(word) for word in text])

sentiment_dict = {
    'excellent': 5,
    'great': 4,
    'good': 3,
    'ok': 2,
    'bad': 1,
    'terrible': 0
}

df['sentiment_score'] = 0

for index, row in df.iterrows():
    for word in row['feedback_text']:
        if word in sentiment_dict:
            df.loc[index, 'sentiment_score'] += sentiment_dict[word]

df.head()

                                               feedback_text                date      sentiment_score