Sentiment Analysis
Walmart Reviews Sentiment Analyzer
Previous Analysis Results
The following table displays the result of sentiment analysis applied on IMEBRANDS product reviews scraped from Walmart marketplace website.
Loading results...
| Product | Review | Rating | Sentiment | Confidence | Categories |
|---|
Sentiment Analysis System
A comprehensive sentiment analysis system designed to analyze Walmart product reviews using both traditional machine learning and advanced Natural Language Processing (NLP) techniques to extract sentiment, categorize feedback, and identify key insights.
Sentiment Analysis
- Multi-class Classification: Categorizes reviews into Negative, Neutral, and Positive
- Random Forest Classifier: Robust model for handling imbalanced datasets
- Confidence Scores: Provides prediction certainty for each classification
Topic Categorization
Customer Service
Product Quality
Shopping Experience
Technical Aspects
Delivery/Store
Value for Money
Data Pipeline
Data Collection
- Kaggle dataset (30,000 reviews)
- Real-time web scraping
Preprocessing
- Text cleaning & normalization
- Tokenization & lemmatization
- Stopword removal
Feature Engineering
- TF-IDF vectorization
- Sentiment scores (TextBlob & VADER)
- Behavioral features
Performance Metrics
82%
Overall Accuracy
0.911
ROC-AUC Score
Class-wise Performance
Positive: 96% | Neutral: 39% | Negative: 78%
Technologies Used
NLTK
Natural Language Toolkit
TextBlob
Text Processing
VADER
Sentiment Analysis
pandas
Data Processing
🤗 Transformers
Deep Learning