Home

Awesome

NewsMTSC: (Multi-)Target-dependent Sentiment Classification in News Articles

NewsMTSC is a dataset for target-dependent sentiment classification (TSC) on news articles reporting on policy issues. The dataset consists of more than 11k labeled sentences, which we sampled from news articles from online US news outlets. More information can be found in our paper published at the EACL 2021.

This repository contains the dataset for target-dependent sentiment classification in news articles reporting on policy issues. Additionally, the repository contains our model named GRU-TSC, which achieves state-of-the-art TSC classification performance on NewsMTSC. Check it out - it works out of the box :-)

Quick start

Reminder: the following description is only relevant if you in fact want to train your own models. If that's not the case, please check above for links to the dataset and our easy-to-use python package.

Installation

It's super easy, we promise! Note that following these instructions is only necessary if you're planning to train a model using our tool. If you only want to predict the sentiment of sentences, please use our Python package, which is even easier to install and use :-)

NewsMTSC was tested on MacOS and Ubuntu; other OS may work, too. Let us know :-)

1. Setup the environment:

This step is optional if you have Python 3.8 installed already (python --version). If you don't have Python 3.8, we recommend using Anaconda for setting up requirements. If you do not have it yet, follow Anaconda's installation instructions.

To setup a Python 3.8 environment (in case you don't have one yet) you may use, for example:

conda create --yes -n newsmtsc python=3.8
conda activate newsmtsc

FYI, for users of virtualenv, the equivalent command would be:

virtualenv -ppython3.8 --setuptools 45 venv
source venv/bin/activate

Note: We recommend Python 3.8, however we have successfully tested NewsMTSC with Python version >=3.8, <3.12.

2. Setup NewsMTSC:

git clone git@github.com:fhamborg/NewsMTSC.git

Afterward, for example, open the project in your IDE and follow the instruction described in the section "Training".

Note that if you only want to classify sentiment using our model, we recommend that you use our PyPI package NewsSentiment. Getting it is as simple as pip install NewsSentiment and using it is four lines of code :-)

Training

If you want to train one of our models or your own model, please clone the repository first.

There are two entry points to the system. train.py is used to train and evaluate a specific model on a specific dataset using specific hyperparameters. We call a single run an experiment. controller.py is used to run multiple experiments automatically. This is for example useful for model selection and evaluating hundreds or thousands of combinations of models, hyperparameters, and datasets.

Running a single experiment

Goal: training a model with a user-defined (hyper)parameter combination.

train.py allows fine-grained control over the training and evaluation process, yet for most command line arguments we provide useful defaults. Two arguments are required:

For more information refer to train.py and combinations_absadata_0.py. If you just want to get started quickly, the command below should work out of the box.

python train.py --own_model_name grutsc --dataset_name newsmtsc-rw

Running multiple experiments

Goal: finding the (hyper)parameter combination to train a model that achieves the best performance.

controller.py takes a set of values for each argument, creates combinations of arguments, applies conditions to remove unnecessary combinations (e.g., some arguments may only be used for a specific model), and creates a multiprocessing pool to run experiments of these argument combinations in parallel. After completion, controller.py creates a summary, which contains detailed results, including evaluation performance, of all experiments. By using createoverview.py, you can export this summary into an Excel spreadsheet.

Support

If you have questions on how to use NewsMTSC or its library, please create a new issue on GitHub. Please understand that we are not able to provide individual support via email. We think that help is more valuable if it is shared publicly so that more people can benefit from it.

Acknowledgements

This repository is in part based on ABSA-PyTorch. We thank Song et al. for making their excellent repository open source.

How to cite

If you use the dataset or model, please cite our paper (PDF):

@InProceedings{Hamborg2021b,
  author    = {Hamborg, Felix and Donnay, Karsten},
  title     = {NewsMTSC: (Multi-)Target-dependent Sentiment Classification in News Articles},
  booktitle = {Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021)},
  year      = {2021},
  month     = {Apr.},
  location  = {Virtual Event},
}