https://github.com/iamdevnd/sentiment-analytics
Real-time Retail & Review Intelligence with Sentiment + Market Basket Analysis
https://github.com/iamdevnd/sentiment-analytics
Last synced: 9 months ago
JSON representation
Real-time Retail & Review Intelligence with Sentiment + Market Basket Analysis
- Host: GitHub
- URL: https://github.com/iamdevnd/sentiment-analytics
- Owner: iamdevnd
- License: mit
- Created: 2025-07-14T20:05:40.000Z (12 months ago)
- Default Branch: main
- Last Pushed: 2025-07-14T20:55:32.000Z (12 months ago)
- Last Synced: 2025-09-18T03:30:23.932Z (9 months ago)
- Language: Jupyter Notebook
- Size: 235 KB
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Sentiment-Analytics
Real-time Retail & Review Intelligence with Sentiment + Market Basket Analysis
# Notebook Description
01_reddit_streaming.ipynb Uses Reddit API (PRAW) to collect live comments from targeted subreddits.
02_cleaning_and_sentiment.ipynb Cleans Reddit data and performs sentiment analysis with TextBlob.
03_amazon_reviews_analysis.ipynb Analyzes review texts for sentiment and links them with star ratings.
04_market_basket_model.ipynb Uses FP-Growth in PySpark for item association rule mining.
# Tech Stack
Language: Python 3.10+
NLP: TextBlob, NLTK
Big Data: PySpark
API & Web: PRAW (Reddit API)
Analysis: Pandas, NumPy, Matplotlib, Seaborn
Modeling: scikit-learn, FP-Growth
# Setup Instructions
# Install Requirements
bash
Copy
Edit
pip install -r requirements.txt
# Reddit API Setup
Go to https://www.reddit.com/prefs/apps
Create a new app and note your:
client_id
client_secret
user_agent
Add them to a .env file:
env
Copy
Edit
REDDIT_CLIENT_ID=your_id
REDDIT_CLIENT_SECRET=your_secret
REDDIT_USER_AGENT=shoplytics_pipeline
# Sample Insights
Positive Reddit sentiment increases on weekends across eCommerce subreddits.
Amazon product reviews with higher sentiment often correlate with 4- and 5-star ratings.
Common itemsets like ['phone_case', 'screen_protector'] have high lift and support, ideal for product bundling.
# Future Enhancements
Integrate Streamlit dashboard for live Reddit sentiment visualization.
Use BERT or DistilBERT for advanced sentiment classification.
Alert system for detecting sentiment dips across product discussions.