Group 4:
- Haodong Zhao
- Somiya Khurram
- Halit Ege Ebiller
Note: We build up the cluster and do some parts of the experiments together, but I commit most parts of the things.
This repo is for the final project of Data Engineering I (1TD169).
The purpose of this project is to develop a horizontally scalable data processing solution to do sentiment analysis: classify Reddit comments as “negative”, “neutral”, or “positive”, by using Natural Language Toolkit(NLTK).
The pipeline of the solution has three steps:
- Inputting Raw Data
- Processing & Classification
- Final results
-
Strong Scaling
-
Weak Scaling
