https://github.com/nrhnnii/social-media-computing
A comprehensive NLP project analyzing Malaysian restaurant reviews using traditional and transformer-based sentiment analysis, social network modeling, and opinion mining techniques.
https://github.com/nrhnnii/social-media-computing
networkx nlp opinion-mining sentiment-analysis spacy text-mining transformers vader
Last synced: about 2 months ago
JSON representation
A comprehensive NLP project analyzing Malaysian restaurant reviews using traditional and transformer-based sentiment analysis, social network modeling, and opinion mining techniques.
- Host: GitHub
- URL: https://github.com/nrhnnii/social-media-computing
- Owner: nrhnnii
- Created: 2025-04-23T08:34:33.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-05-27T03:45:34.000Z (about 1 year ago)
- Last Synced: 2025-05-27T04:33:30.927Z (about 1 year ago)
- Topics: networkx, nlp, opinion-mining, sentiment-analysis, spacy, text-mining, transformers, vader
- Language: Jupyter Notebook
- Homepage:
- Size: 3.11 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# 📚 Natural Language Processing on Malaysian Restaurant Reviews
This repository explores a full NLP pipeline using real-world customer reviews from Google Maps. The tutorials cover everything from basic preprocessing to advanced sentiment analysis and social network modeling — all using the same dataset:
📁 **Dataset**: [GoogleReview_data_cleaned.csv](https://www.kaggle.com/datasets/choonkhonng/malaysia-restaurant-review-datasets)
🗂️ **Source**: Google Maps reviews of restaurants across Malaysia
---
## 🔍 Tutorials Overview
### 📁 Tutorial 3 & 4 – Text Preprocessing & NLP with NLTK
**Goal**: Learn essential text preprocessing techniques
**Topics**:
- Sentence segmentation, tokenization, case folding
- Punctuation and stopword removal
- POS tagging, stemming
- Named Entity Recognition (NER) using spaCy
- Word cloud visualization
---
### 📁 Tutorial 5 & 6 – Sentiment Analysis with TextBlob & VADER
**Goal**: Extract sentiment orientation using lexicon-based models
**Tools**: TextBlob, VADER
**Tasks**:
- Polarity and subjectivity scoring
- Social-media-aware sentiment analysis
- Comparing and visualizing both tools
---
### 📁 Tutorial 7 – Social Network Analysis with NetworkX & Louvain
**Goal**: Discover user behavior through reviewer–restaurant relationships
**Tools**: NetworkX, community-louvain
**Tasks**:
- Bipartite graph modeling
- Reviewer similarity projection
- Community detection and visualization
---
### 📁 Tutorial 8 & 9 – Opinion Mining & Sentiment Enrichment with Transformers
**Goal**: Go deeper into meaning, tone, and nuance in user reviews
**Tools**: spaCy, pywsd, Hugging Face Transformers, VADER
**Tasks**:
- Named Entity Recognition (NER)
- Word Sense Disambiguation (WSD)
- Sentence sentiment classification (RoBERTa)
- Lexicon-based word sentiment
- Polarity and sentiment strength extraction