ABSA-datasets

datasets for Aspect-Based Sentiment Analysis and codes for reading them.

goal

1. Try to collect all ABSA datasets
2. Provide uniform interface to read the datasets
3. Since there are no official train and development splits for most ABSA datasets, we try to provide standard splits for them

Supported datasets

SemEval-2014 Task 4

1. SemEval-2014-Task-4-LAPT
2. SemEval-2014-Task-4-REST

SemEval-2015 Task 12

1. SemEval-2015-Task-12-LAPT
2. SemEval-2015-Task-12-REST
3. SemEval-2015-Task-12-HOTEL

SemEval-2016 Task 5

1. SemEval-2016-Task-5-CH-CAME-SB1
2. SemEval-2016-Task-5-CH-PHNS-SB1
3. SemEval-2016-Task-5-LAPT-SB1
4. SemEval-2016-Task-5-LAPT-SB2
5. SemEval-2016-Task-5-REST-SB1
6. SemEval-2016-Task-5-REST-SB2

bdci2019

1. bdci2019-internet-news-sa
2. bdci2019-financial-negative

A Challenge Dataset and Effective Models for Aspect-Based Sentiment Analysis

1. MAMSACSA
2. MAMSATSA

nlpcc2012

nlpcc2012-weibo-sa

Usage

from data_adapter.data_object import get_dataset_class_by_name

dataset_name = 'SemEval-2014-Task-4-REST'
dataset = get_dataset_class_by_name(dataset_name)()

Related Repositories

l294265421/ACSA

Contributions

Feel free to contribute!

You can raise an issue or submit a pull request, whichever is more convenient for you.

Licence

MIT License