Yelp Restaurant Sentiment Analysis Project
Description
In this project, I leveraged machine learning and natural language processing (NLP) techniques to analyze large-scale Yelp restaurant reviews, extracting valuable insights into customer sentiment and business performance.
Project Highlights:
-
Optimized data preprocessing pipelines by handling missing values, performing text normalization, and applying memory-efficient transformations to enhance sentiment classification accuracy.
-
Performed geospatial clustering using K-Means and DBSCAN, identifying high-performing restaurant clusters with an average rating of 4.13 stars and 1,600+ reviews per business.
-
Conducted sentiment analysis on massive Yelp review datasets, uncovering a 0.66 correlation between sentiment scores and star ratings for pizza restaurants, highlighting key factors influencing customer satisfaction.
-
Developed regional performance insights, enabling restaurant owners and marketing teams to refine business strategies based on geographical trends and customer sentiment patterns.
Project Impact:
This project provides data-driven recommendations for restaurant owners, helping them understand customer sentiment, optimize marketing strategies, and improve service quality. By leveraging geospatial clustering and sentiment analysis, the findings offer actionable insights to identify competitive advantages and enhance customer experience in various locations.
Skills
- Python: Pandas, NumPy, Scikit-Learn, NLTK, TextBlob
- Machine Learning: Clustering: K-means, DBSCAN, Sentiment Analysis
- Visualization Tools: Tableau, Matplotlib, Seaborn
- GeoSpatial Analysis: Geopandas, Folium
