tawsifsazid/Unified-Representation-for-Argumentation-Mining

Novel unified representation to solve all the sub-tasks of argumentation mining

• Argumentation Mining

  

• Paper links

  • A Unified Representation and a Decoupled Deep Learning Architecture for Argumentation Mining of Students' Persuasive Essays (AM_2022_coling2022)

    We develop a novel unified representation for the argumentation mining task facilitating the extracting from text and the labelling of the non-argumentative units and argumentation components---premises, claims, and major claims---and the argumentative relations---premise to claim or premise in a support or attack relation, and claim to major-claim in a for or against relation---in an end-to-end machine learning pipeline. This tightly integrated representation combines the component and relation identification sub-problems and enables a unitary solution for detecting argumentation structures. This new representation together with a new deep learning architecture composed of a mixed embedding method, a multi-head attention layer, two biLSTM layers, and a final linear layer obtain state-of-the-art accuracy on the Persuasive Essays dataset. Also, we have introduced a decoupled solution to identify the entities and relations first, and on top of that a second model is used to detect distance between the detected related components. An augmentation of the corpus (paragraph version) by including copies of major claims has further increased the performance.

  • A Unified Representation and Deep Learning Architecture for Argumentation Mining of Students' Persuasive Essays (CMNA22)

    We develop a novel unified representation for the argumentation mining task facilitating the extracting from text and the labelling of the non-argumentative units and argumentation components---premises, claims, and major claims---and the argumentative relations---premise to claim or premise in a support or attack relation, and claim to major-claim in a for or against relation---in an end-to-end machine learning pipeline. This tightly integrated representation combines the component and relation identification sub-problems and enables a unitary solution for detecting argumentation structures. This new representation together with a new deep learning architecture composed of a mixed embedding method, a multi-head attention layer, two biLSTM layers, and a final linear layer obtain state-of-the-art accuracy on the Persuasive Essays dataset. An augmentation of the corpus (Paragraph version) by including copies of major claims has further increased the performance.

• Unified Representation for Argumentation Mining

  

• Model_Architecture (Both Unfied_AM and Decoupled_AM (Unfied_AM + 2nd model to detect distance))

  

• Notes:

  • New augmented corpus are available in the datasets folder.

  • The datasets (path: datasets/original) that we have used for detecting argument structures are taken from https://github.com/UKPLab/acl2017-neural_end2end_am/tree/master/data/conll

• To reproduce the results from the papers:

  1. Clone the repo: git clone https://github.com/tawsifsazid/Unified-Representation-for-Argumentation-Mining.git

  2. Create an environment (Python 3.9.7) and run: pip install -r requirements.txt

  3. Install PyTorch: conda install pytorch torchvision torchaudio pytorch-cuda=11.6 -c pytorch -c nvidia or from the official website

  4. The weight files can be downloaded from the links. Folder name 'weight' should contain the downloaded files before running the code.

     • unified_am model weight file for augmented_paragraph corpus

     • unified_am model weight file for original_essay corpus

     • unified_am model weight file for original_paragraph corpus

  5. Run python main.py to reproduce the results

  6. Default config is set for original essay level corpus.

• Cite:

  • @inproceedings{sazid-mercer-2022-unified, title = "A Unified Representation and a Decoupled Deep Learning Architecture for Argumentation Mining of Students{'} Persuasive Essays", author = "Sazid, Muhammad Tawsif and Mercer, Robert E.", booktitle = "Proceedings of the 9th Workshop on Argument Mining", month = oct, year = "2022", address = "Online and in Gyeongju, Republic of Korea", publisher = "International Conference on Computational Linguistics", url = "https://aclanthology.org/2022.argmining-1.6", pages = "74--83", abstract = "We develop a novel unified representation for the argumentation mining task facilitating the extracting from text and the labelling of the non-argumentative units and argumentation components{---}premises, claims, and major claims{---}and the argumentative relations{---}premise to claim or premise in a support or attack relation, and claim to major-claim in a for or against relation{---}in an end-to-end machine learning pipeline. This tightly integrated representation combines the component and relation identification sub-problems and enables a unitary solution for detecting argumentation structures. This new representation together with a new deep learning architecture composed of a mixed embedding method, a multi-head attention layer, two biLSTM layers, and a final linear layer obtain state-of-the-art accuracy on the Persuasive Essays dataset. Also, we have introduced a decoupled solution to identify the entities and relations first, and on top of that, a second model is used to detect distance between the detected related components. An augmentation of the corpus (paragraph version) by including copies of major claims has further increased the performance.", }

  • @inproceedings{sazid2022unified, title={A unified representation and deep learning architecture for argumentation mining of students’ persuasive essays}, author={Sazid, Muhammad Tawsif and Mercer, Robert E}, booktitle={to appear, CEUR Workshop Proceedings}, year={2022} }