arusl / anlp_nlp2022_a6-1

This repository contains codes related to the experiments in "On the Applicability of Zero-Shot Cross-Lingual Transfer Learning for Sentiment Classification in Distant Language Pairs".

Overview

This repository contains codes related to the experiments in the following research paper.

Paper Title: On the Applicability of Zero-Shot Cross-Lingual Transfer Learning for Sentiment Classification in Distant Language Pairs (PDF)

Authors: Andre Rusli and Makoto Shishido (Tokyo Denki University)

which was presented at the 28th Annual Meeting of the Association of Natural Language Processing (言語処理学会第28回年次大会(NLP2022))

Conference website: https://www.anlp.jp/nlp2022/index.html.

Datasets

The dataset used in the experiments can be downloaded from the below links, and should be put under a directory named data/:

• amazon-enja-sentiment-dataset/ (source)
• rakuten-sentiment-dataset/ (source)
• indonesian-sentiment-dataset/
  • indolem/ (source)
  • indonlu/ (source)

Model Files

The models that were trained during the experiments can be downloaded from the below links, and should be put under a directory named models/:

• en-binary-model/
• ja-binary-model/
• enja-binary-model/