https://github.com/oya163/nepali-stemmer
Simple rule-based Nepali stemmer. Flask web app deployed on Heroku platform. Created pip package.
https://github.com/oya163/nepali-stemmer
flask heroku linguistics nepali-dictionary nepali-stemmer pip stemmer
Last synced: 4 months ago
JSON representation
Simple rule-based Nepali stemmer. Flask web app deployed on Heroku platform. Created pip package.
- Host: GitHub
- URL: https://github.com/oya163/nepali-stemmer
- Owner: oya163
- License: mit
- Created: 2020-04-23T20:38:57.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2020-04-26T03:44:18.000Z (about 5 years ago)
- Last Synced: 2025-01-16T06:17:10.829Z (5 months ago)
- Topics: flask, heroku, linguistics, nepali-dictionary, nepali-stemmer, pip, stemmer
- Language: Python
- Homepage: https://nepali-stemmer.herokuapp.com
- Size: 167 KB
- Stars: 3
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Nepali Stemmer
This is a simple Nepali stemmer. It iteratively separates out the suffixes (postpositions) until no more separation can be processed. The algorithm is based on hindi-stemmer.
## Features:
- Iterative separation
- Handles the postposition attached with punctuations carefully
- Example: नेपाललाई, -> नेपाल लाई,
- Basic text cleaning
- Cross-verification with Nepali dictionary
## How to install
pip install -r requirements.txt
pip install nepali-stemmer
## How to run
from nepali_stemmer.stemmer import NepStemmer
nepstem = NepStemmer()
nepstem.stem("नेपालको एमाले पार्टीका झोले, मण्डलेहरु अमेरिका आउने रे !")
'नेपाल को एमाले पार्टी का झोले, मण्डले हरु अमेरिका आउने रे !'
## Deployment
A simple flask based web app [deployment](https://nepali-stemmer.herokuapp.com/)
## To-do:
- [ ] Word transformation with stemming process
- [ ] IR evaluation
- [ ] Code-mixed data
## References:
- Suffix list: https://github.com/birat-bade/NepaliStemmer
- Nepali Dictionary : https://github.com/PraveshKoirala/stemmer
- Algorithm : https://github.com/sainimohit23/hindi-stemmer
## Contact
Email: [oyashi](mailto:[email protected])
Note: Project created during COVID-19 quarantine out-of-boredom and necessity