• Corpus ID: 3339724

Cross-topic Argument Mining from Heterogeneous Sources Using Attention-based Neural Networks

@article{Stab2018CrosstopicAM,
  title={Cross-topic Argument Mining from Heterogeneous Sources Using Attention-based Neural Networks},
  author={Christian Stab and Tristan Miller and Iryna Gurevych},
  journal={ArXiv},
  year={2018},
  volume={abs/1802.05758}
}
Argument mining is a core technology for automating argument search in large document collections. [] Key Method We source annotations for over 25,000 instances covering eight controversial topics. The results of cross-topic experiments show that our attention-based neural network generalizes best to unseen topics and outperforms vanilla BiLSTM models by 6% in accuracy and 11% in F-score.
View PDF on arXiv
View With Semantic Reader
29 Citations
Highly Influential Citations
8
Background Citations
14
Methods Citations
13
Results Citations
4

Figures and Tables from this paper

29 Citations

Citation Type

Citation Type

Lexicon Guided Attentive Neural Network Model for Argument Mining
TLDR
This paper proposes a methodology to integrate lexicon information into a neural network model by attention mechanism and conducts experiments on the UKP dataset, which is collected from heterogeneous sources and contains several text types, e.g., microblog, Wikipedia, and news.
Using Convolutional Neural Network in Cross-Domain Argumentation Mining Framework
TLDR
This paper presents a new framework for Argumentation Mining to detect argumentative segments and their components automatically using Convolutional Neural Network (CNN) and shows challenging results in comparison with classic machine learning models.
Towards an argumentative content search engine using weak supervision
TLDR
This work uses a weak signal to define weak signals for training DNNs to obtain significantly greater performance, and adapts the system to solve a recent argument mining task of identifying argumentative sentences in Web texts retrieved from heterogeneous sources, and obtain F1 scores comparable to the supervised baseline.
TACAM: Topic And Context Aware Argument Mining
TLDR
This work argues that topic information is crucial for argument mining, since the topic defines the semantic context of an argument, and proposes different models for the classification of arguments, which take information about a topic of a argument into account.
Cross-Lingual Argumentation Mining
TLDR
This document presents an argumentation mining system to extract arguments from news articles or essays, in Portuguese and English, using a cross-language machine learning approach based on deep neural networks.
Corpus Wide Argument Mining - a Working Solution
TLDR
This work presents a first end-to-end high-precision, corpus-wide argument mining system, made possible by combining sentence-level queries over an appropriate indexing of a very large corpus of newspaper articles, with an iterative annotation scheme.
ArgumenText: Searching for Arguments in Heterogeneous Sources
TLDR
This paper presents an argument retrieval system capable of retrieving sentential arguments for any given controversial topic, and finds that its system covers 89% of arguments found in expert-curated lists of arguments from an online debate portal, and also identifies additional valid arguments.
Active Learning for Argument Mining: A Practical Approach
TLDR
In a large scale comparison of several Active Learning methods, it is shown that Active Learning considerably decreases the effort necessary to get good deep learning performance on the task of Argument Unit Recognition and Classification (AURC).
Semi-supervised Cleansing of Web-based Argument Corpora
TLDR
This thesis introduces an approach to automatically filter unwanted segments of the argument texts based on a small predefined set of ground truth patterns, which automatically searches and classifies sentences according to their importance towards an argument.
Cross-Lingual Argumentative Relation Identification: from English to Portuguese
TLDR
This paper presents the first attempt to address the task of argumentative relation identification, a subtask of argument mining, in a cross-lingual setting, and compares two standard strategies for cross-language learning, namely: projection and direct-transfer.
...
1
2
3
...

References

SHOWING 1-10 OF 36 REFERENCES
Cross-Domain Mining of Argumentative Text through Distant Supervision
TLDR
A distant supervision approach that acquires argumentative text segments automatically from online debate portals and freely provides the underlying corpus for research, showing that training on such a corpus improves the effectiveness and robustness of mining argumentativeText.
Understanding and Detecting Diverse Supporting Arguments on Controversial Issues
TLDR
Experimental results show that LambdaMART (Burges, 2010) ranker that uses features informed by argument types yields better performance than the same ranker trained without type information.
Neural End-to-End Learning for Computational Argumentation Mining
TLDR
This work investigates neural techniques for end-to-end computational argumentation mining and finds that jointly learning 'natural' subtasks, in a multi-task learning setup, improves performance.
Context Dependent Claim Detection
TLDR
This work formally defines the challenging task of automatic claim detection in a given context and outlines a preliminary solution, and assess its performance over annotated real world data, collected specifically for that purpose over hundreds of Wikipedia articles.
What is the Essence of a Claim? Cross-Domain Claim Identification
TLDR
While the divergent conceptualization of claims in different datasets is indeed harmful to cross-domain classification, it is shown that there are shared properties on the lexical level as well as system configurations that can help to overcome these gaps.
Discourse Parsing with Attention-based Hierarchical Neural Networks
TLDR
Experimental results show that the proposed attention-based hierarchical neural network model for discourse parsing obtains comparable performance to the contemporary state-of-the-art systems with little manual feature engineering.
Context-aware Argumentative Relation Mining
TLDR
This paper presents contextaware argumentative relation mining that uses features extracted from writing topics as well as from windows of context sentences that improve predictive performance in two argumentative relations classification tasks.
Argumentation Mining on the Web from Information Seeking Perspective
TLDR
It is argued that an annotation scheme for argumentation mining is a function of the task requirements and the corpus properties and it is found that the choice of the argument components to be annotated strongly depends on the register, the length of the document, and inherently on the literary devices and structures used for expressing argumentation.
Argument Extraction from News, Blogs, and Social Media
TLDR
The proposed approach tries to identify the exact fragments that contain the premises from the sentences that contain arguments, by utilizing conditional random fields, and according to literature, are quite promising.
Improved Representation Learning for Question Answer Matching
TLDR
This work develops hybrid models that process the text using both convolutional and recurrent neural networks, combining the merits on extracting linguistic information from both structures to address passage answer selection.
...
1
2
3
4
...