Skip to content
/ Amazon-Product-Review-Sentiment-Classification Public

Build end-to-end Data Science (NLP) project for Amazon's Product Review Sentiment Classification using AWS tools.

0 stars 0 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

rkshiyaniya/Amazon-Product-Review-Sentiment-Classification

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
Detect data bias with Amazon SageMaker Clarify.ipynb
Detect data bias with Amazon SageMaker Clarify.ipynb
 
 
Ingest, Transform and Visualize Data with AWS.ipynb
Ingest, Transform and Visualize Data with AWS.ipynb
 
 
README.md
README.md
 
 
avg_sentiment_per_category.png
avg_sentiment_per_category.png
 
 
balanced_report.pdf
balanced_report.pdf
 
 
distribution_num_words_per_review.png
distribution_num_words_per_review.png
 
 
distribution_sentiment_per_category.png
distribution_sentiment_per_category.png
 
 
num_reviews_per_category.png
num_reviews_per_category.png
 
 
unbalanced_report.pdf
unbalanced_report.pdf
 
 
womens_clothing_ecommerce_reviews.csv
womens_clothing_ecommerce_reviews.csv
 
 
womens_clothing_ecommerce_reviews_balanced.csv
womens_clothing_ecommerce_reviews_balanced.csv
 
 
womens_clothing_ecommerce_reviews_transformed.csv
womens_clothing_ecommerce_reviews_transformed.csv
 
 

Repository files navigation

Amazon's Product Review Sentiment Classification

Ingest, Transform & Visualize Data with AWS

Here's some Screenshots of Analysis

Number Of Reviews per Category

num_reviews_per_category

Distribution of Number Of Words per Review

distribution_num_words_per_review

Average Sentiment per Category

avg_sentiment_per_category

Distribution of Sentiment per Category

distribution_sentiment_per_category

Raw Data

Transformed Data

Statistical Bias Detection in Dataset and Balanced it

Balanced Data

Report Of Unbalanced Data

Report Of Balanced Data

  • Built model using AutoML (FastText algorithm) with Amazon SageMaker Autopilot
  • Train Model using built-in BlazingText & BERT algorithms

About

Build end-to-end Data Science (NLP) project for Amazon's Product Review Sentiment Classification using AWS tools.

Topics

visualization python nlp aws machine-learning sql sentiment-analysis pytorch fasttext bert automl amazon-s3 amazon-athena model-deployment multi-class-classification amazon-sagemaker blazingtext amazon-sagemaker-studio built-in-algorithms statistical-bias-detection

Resources

Readme
Activity

Stars

0 stars

Watchers

1 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages