Flipkart-review-scraping

This is a project on scraping user reviews for a perticular product in flipkart india. The program scrape the url of the review section of a perticular project and gives out two csv file with negative reviews and positive review with its respective word-cloud to make a better understanding on what the reviews are about.

Overall function :

The python program uses requests and BeautifulSoup liberaries The program srapes flipkart's review page and wirtes in csv file Reviews are seperated as positive and negative. The data contains:

Star rating

Review title

Review comment

creating a review table

Generating a world cloud for both Positive and Negative reviews

Generates a Bar Graph for displaying Overall star ratings.

Manual changes you might need :

1. The url link must be the "All reviews" page of a perticular product.
2. Make sure the "class-name" or "id" match the page's html script you're trying to scrape.
3. Make sure you have installed all the modules used in program.

Note

flipkart usually has the same classes and ID on html tags of html. But if the program shows an error, make sure your url has the same class-name and id in the html page.

That is :

    paragraphs = soup.find_all('p', class_='z9E0IG')
    stars = soup.find_all(class_='XQDdHH Ga3i8K')
    comments = soup.find_all(class_='ZmyHeo')

Here, make sure to check with the class code in the url you have incase if the program failed to sc

Note

The Required modules are : - Requests - Bs4 (beautifulSoup) - csv - matplotlib - wordcloud - plotly