https://github.com/andreaschandra/ugm-aes

https://github.com/andreaschandra/ugm-aes

Last synced: about 1 year ago
JSON representation

• Host: GitHub
• URL: https://github.com/andreaschandra/ugm-aes
• Owner: andreaschandra
• Created: 2019-09-30T14:06:53.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2019-09-30T14:58:27.000Z (over 6 years ago)
• Last Synced: 2025-02-12T09:39:17.238Z (over 1 year ago)
• Language: Jupyter Notebook
• Size: 14.6 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# UGM UKARA 1.0 Challenge

## Task Definition

    

    Research in the area of automatic short-answer scoring is getting more and more attention 

    due to the need of making the assessment process faster. Despite the rapid progress on 

    automatic short-answer assessment, research in this area particularly for Bahasa Indonesia 

    has been very limited and only recently emerged as a topic.

    

    The main challenges on the development of automatic short-answer grading 

    for Bahasa Indonesia is the use of informal language in writing context. 

    The diversity of local language of the student is one of many factors that influence the writing 

    style and the vocabulary used by the students. In addition, students often include slang words

    which commonly used in daily conversation. 

    

    UKARA 1.0 Challenge aims to encourage more ideas and studies for developing automatic 

    short-answer scoring specifically for Bahasa Indonesia.

    

## Timeline

Event | Date

---|---

Challenge launch | July 29, 2019

Dev phase submission | July 29, 2019 – Sept 10, 2019

Test dataset begins | September 16, 2019

Submission close | September 19, 2019

Result announcement | September 20, 2019

Working paper | October 7, 2019

Workshop and Winners announcement | October 14, 2019

## Problem

    

    This competition examines model on:

        - low resource

        - out of vocabulary

        - a lot of words contain prefix and suffix

        - 

    

## Goal

    

    Win this competition

    

## Folder Trees

    |-- data

        |-- raw

        |-- cleansed

        |-- feature-representation

        |-- support

    |-- notebook

    |-- model

    |-- readme.md

    |-- requirements.txt

    

## FAQ

Please install libraries that used in this notebook

    

    pip install -r requirements.txt

    

Data can be found at [nlp-ugm](https://nlp.mipa.ugm.ac.id/ukaratrack1/)

All experiments can be found [here](https://docs.google.com/spreadsheets/d/1vXnHC-twrSLsdpZoPP4UvzHiCjAb-krYZF531kK77k0/edit?usp=sharing)

Indonesian Word Embeeding can be found [here](https://medium.com/@diekanugraha/membuat-model-word2vec-bahasa-indonesia-dari-wikipedia-menggunakan-gensim-e5745b98714d)