Skip to content

Webscraped flipkart realme 5 pro reviews using beautifulsoup package and did NLP to predict Ratings

Notifications You must be signed in to change notification settings

kkarthikvk/WebScrapping-NLP

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 

Repository files navigation

WebScrapping-NLP

Webscraped flipkart realme 5 pro reviews using beautifulsoup package and did NLP to predict Ratings Find the Co-Relation between Ratings and Reviews BRIEFLY: 1.From flipkart page i have taken the reviews of realme 5 pro phone . 2.In which each review is in one page so we have to take the url separately for each page and parser each page . 3.After parsing into the page . Collect all the information in one variable and access the information by finding the root class. 4.After finding the root class . Using loop statement try to find the information hidden inside the class by giving index value to the stored variable. 5.Then collect all the data that are all necessary i have scraped reviews and rating to find the co-relation between them . 6. We do all preprocessing steps to convert sentence into numeric , So that our model could perform. 7.NLTK preprocessing steps:- (i). Tokenization using re (ii).Converting into lowercase (iii).Removing numbers and Punctuations using string (iv).Stemming using PorterStemmer (v). Vectorization using sklearn.TfIdfvectorization 8.Used GradientBoostingClassifier to predict the Ratings

NOTE:- only scraped 3 pages and trained the model with less number of data . So , got accuracy of 99% for training and 67% for testing set . If we have more data we could get more accuracy.

Added via terminal

About

Webscraped flipkart realme 5 pro reviews using beautifulsoup package and did NLP to predict Ratings

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published