Back to projects
Sentiment Analysis at Scale
Sentiment classifier built in PySpark to process a large text dataset and evaluate text-classification performance.
- Project
- MSc Machine Learning on Big Data module
- Role
- NLP & big data
- Period
- 2026
[ Overview ]
The problem
Classifying sentiment across a large text dataset requires distributed processing.
What I built
Built the preprocessing and classification pipeline in PySpark and evaluated its performance.
[ Highlights ]
- Distributed text preprocessing in PySpark
- Sentiment classification model
- Evaluation of text-classification performance
[ Stack ]
- Python
- PySpark
- NLP