Sentiment Analysis from Google, Zomato, Swiggy & Yelp – Technical Tutorial

Introduction

Customer sentiment is one of the strongest indicators of:

• brand loyalty
• product experience
• operational performance
• food quality & delivery issues
• restaurant hygiene perceptions
• rider behavior trends
• neighborhood sentiment footprint

When businesses operate across:

• multiple delivery platforms
• multiple regions
• hundreds of stores

...they need automated review extraction + sentiment modeling to understand what customers really feel.

This tutorial shows how to build a full Sentiment Intelligence Pipeline using:

• Python
• Selenium
• BeautifulSoup
• NLP (TextBlob / Vader / keyword scoring)
• Multi-platform review ingestion

Platforms covered:

• Google Maps
• Yelp
• Zomato
• Swiggy

This is the same approach Actowiz Solutions uses in enterprise-grade sentiment dashboards.

How Do You Set Up the Environment for Review Extraction?

Install the required packages:

pip install selenium
pip install undetected-chromedriver
pip install beautifulsoup4
pip install pandas
pip install textblob
pip install nltk
pip install lxml

These provide:

• web automation
• HTML parsing
• sentiment classification
• data handling

Import everything:

import undetected_chromedriver as uc
from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
from bs4 import BeautifulSoup
from time import sleep
import pandas as pd
from textblob import TextBlob
import re

What Data Will You Extract From Each Platform?

The scraper collects:

• review text
• rating
• review date
• reviewer name (if available)
• source platform
• store or restaurant name
• location (Google-specific)

Sample output:

{
  "platform": "Google",
  "store_name": "Starbucks, Times Square",
  "rating": 4,
  "review_text": "Loved the place! Quick service.",
  "date": "2 weeks ago",
  "sentiment": 0.75
}

How Do You Scrape Reviews From Google Maps?

Google Maps loads review text inside a scrollable container.

1. Open Google Maps Listing

browser = uc.Chrome()
browser.get("https://www.google.com/maps/place/Starbucks+Times+Square")
sleep(4)

2. Open the Review Panel

review_btn = browser.find_element(By.XPATH, '//button[contains(@aria-label,"reviews")]')
review_btn.click()
sleep(3)

3. Scroll the Review Container

panel = browser.find_element(By.CLASS_NAME, "m6QErb")

for _ in range(40):
    browser.execute_script("arguments[0].scrollTop = arguments[0].scrollHeight", panel)
    sleep(1)

4. Extract Review Cards

gm_reviews = []
cards = browser.find_elements(By.XPATH, '//div[@data-review-id]')

5. Parse Google Reviews

for c in cards:
    try:
        author = c.find_element(By.CLASS_NAME, "d4r55").text
    except:
        author = ""

    try:
        rating = c.find_element(By.CLASS_NAME, "fzvQZe").get_attribute("aria-label")
    except:
        rating = ""

    try:
        text = c.find_element(By.CLASS_NAME, "wiI7pd").text
    except:
        text = ""

    try:
        date = c.find_element(By.CLASS_NAME, "rsqaWe").text
    except:
        date = ""

    gm_reviews.append({
        "platform": "Google",
        "author": author,
        "rating": rating,
        "review_text": text,
        "date": date
    })

How Do You Scrape Reviews From Yelp?

1. Open Yelp Listing

browser.get("https://www.yelp.com/biz/starbucks-new-york")
sleep(4)

2. Scroll to Load Reviews

for _ in range(20):
    browser.find_element(By.TAG_NAME, "body").send_keys(Keys.END)
    sleep(1)

3. Extract Yelp Review Blocks

yelp_reviews = []
blocks = browser.find_elements(By.XPATH, '//li[contains(@class,"review")]')

4. Parse Yelp Reviews

for b in blocks:
    try:
        text = b.find_element(By.CLASS_NAME, "comment").text
    except:
        text = ""

    try:
        rating = b.find_element(By.XPATH, './/div[contains(@aria-label,"star rating")]').get_attribute("aria-label")
    except:
        rating = ""

    try:
        date = b.find_element(By.CLASS_NAME, "css-chan6m").text
    except:
        date = ""

    yelp_reviews.append({
        "platform": "Yelp",
        "rating": rating,
        "review_text": text,
        "date": date
    })

How Do You Scrape Reviews From Zomato?

Zomato uses dynamic containers and pagination.

1. Open Zomato Listing

browser.get("https://www.zomato.com/mumbai/starbucks-fort/reviews")
sleep(4)

2. Scroll Multiple Times

for _ in range(20):
    browser.find_element(By.TAG_NAME, "body").send_keys(Keys.END)
    sleep(1.2)

3. Extract Review Blocks

zo_reviews = []
blocks = browser.find_elements(By.XPATH, '//div[contains(@class,"rev-text")]')

4. Parse Zomato Reviews

for b in blocks:
    try:
        text = b.text
    except:
        text = ""

    # Rating (inside preceding sibling div)
    try:
        rating = b.find_element(By.XPATH, '../../div/div[contains(@class,"rated")]').text
    except:
        rating = ""

    zo_reviews.append({
        "platform": "Zomato",
        "rating": rating,
        "review_text": text
    })

How Do You Scrape Reviews From Swiggy?

Swiggy reviews appear under store pages.

1. Open Swiggy Store URL

browser.get("https://www.swiggy.com/restaurants/starbucks-bandra-west-mumbai")
sleep(4)

2. Scroll to Load Reviews

for _ in range(20):
    browser.find_element(By.TAG_NAME, "body").send_keys(Keys.END)
    sleep(1)

3. Extract Review Blocks

sw_reviews = []

blocks = browser.find_elements(By.XPATH, '//div[contains(@class,"_2xsFi")]')

4. Parse Swiggy Reviews

for b in blocks:
    try: text = b.find_element(By.CLASS_NAME, "_2xsFi").text
    except: text = ""

    try:
        rating = b.find_element(By.CLASS_NAME, "_3LWZl").text
    except:
        rating = ""

    sw_reviews.append({
        "platform": "Swiggy",
        "rating": rating,
        "review_text": text
    })

How Do You Combine Reviews From All Platforms?

df = pd.DataFrame(gm_reviews + yelp_reviews + zo_reviews + sw_reviews)

How Do You Clean Different Rating Formats?

def extract_rating(val):
    nums = re.findall(r"\d+\.?\d*", str(val))
    return float(nums[0]) if nums else None

df["rating_num"] = df["rating"].apply(extract_rating)

How Do You Perform Sentiment Analysis?

We use TextBlob for polarity scores:

def get_sentiment(text):
    return TextBlob(text).sentiment.polarity

df["sentiment"] = df["review_text"].apply(get_sentiment)

Values:

• > 0 = Positive
• 0 = Neutral
• < 0 = Negative

How Do You Build Platform-Wise Sentiment Scores?

df.groupby("platform")["sentiment"].mean()

How Do You Identify Complaint Themes Automatically?

Define keyword groups:

complaints = {
    "service": ["slow", "rude", "unprofessional"],
    "delivery": ["late", "cold", "leaking", "delay"],
    "taste": ["bland", "bad taste", "not good"],
    "cleanliness": ["dirty", "messy", "unclean"],
    "pricing": ["expensive", "overpriced"]
}

Tag each review:

for tag, words in complaints.items():
    df[f"kw_{tag}"] = df["review_text"].apply(
        lambda x: any(w.lower() in x.lower() for w in words)
    )

How Do You Export the Sentiment Intelligence Dataset?

df.to_csv("sentiment_analysis_reviews.csv", index=False)

What Are the Limitations of This Scraper?

• DOM structures change frequently on Zomato & Swiggy.
• Google Maps throttles rapidly if scrolling is too fast.
• Yelp uses aggressive anti-bot systems → proxies required.
• Sentiment models might misinterpret sarcasm.
• Language diversity across reviews reduces accuracy.

When Should You Use Actowiz Solutions Instead of Custom Scripts?

DIY scripts work if:

• you need a few hundred reviews
• one-time extraction
• small geographic coverage

But Actowiz Solutions is better for:

• 1–10 million reviews per month
• multi-country scraping
• daily sentiment dashboards
• automated complaint tagging
• competitor-wise review monitoring
• platform-level comparison (Google vs Zomato vs Swiggy vs Yelp)
• enterprise compliance + clean formatting

We deliver:

• APIs
• CSV/JSON feeds
• custom dashboards
• NLP-based insights
• alert systems for negative spikes

Conclusion

In this tutorial, you learned how to:

• Extract reviews from Google Maps, Yelp, Zomato, and Swiggy
• Normalize and clean ratings
• Run sentiment analysis
• Identify complaint themes
• Build cross-platform intelligence datasets
• Export insights for dashboards
• Scale the pipeline for large hospitality & retail brands

This forms the backbone of Actowiz Solutions’ Sentiment Intelligence for F&B, retail, delivery, and hospitality brands.

You can also reach us for all your mobile app scraping, data collection, web scraping , and instant data scraper service requirements!