BTS Text Classifier to determine train service status
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
Example usage.ipynb
README.md
bts_classifier.py
model.pkl
requirements.txt

README.md

BTS SkyTrain text classifier

This repository provides a Python-based Machine Learning model for classificating BTS SkyTrain's status from its Twitter text.

Installation

pip install -r requirements.txt

Which, to be honest, only scikit-learn is needed.

Usage

>>> import bts_classifier
>>> tweet_text = "20.40 ขบวนรถไฟฟ้าขัดข้องที่สถานีบางหว้า ขบวนรถในสายสีลมจะล่าช้า ขออภัยในความไม่สะดวก"
>>> status = bts_classifier.classify(tweet_text)
>>> status
'disrupted'

The status after the classification can be any of these four values:

  • None (with NoneType) indicates that the processed tweet is unrelated to train service status
  • 'normal' indicates that the train had returned to its normal service
  • 'disrupted' indicates that the train has paused its operation.
  • 'delay' indicates that although there is the train service, but passengers are warned to expect train delays.