Computing controversy: Formal model and algorithms for detecting controversy on Wikipedia and in search queries

Zieliński, K; Nielek, Radosław; Wierzbicki, Adam; Jatowt, Adam

doi:10.1016/j.ipm.2017.08.005

Cited by 20 publications

(23 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As stated by Zielinski et al (2018), detecting controversy and controversial themes in social media through automatic methods is especially important, since presenting users with the indications and explanations of the controversy generated by the content they consume allows them to see the "wider picture" instead of leading them to obtain one-sided views. The authors summarize how controversy has been explored from multiple lenses, including social science, traditional media, social media, and Web search.…”

Section: Controversy In Online Discussionmentioning

confidence: 99%

“…For example, Dori-Hacohen and Allan (2015) measured controversy of web pages mapping these pages to Wikipedia articles, in which topic controversy can be measured. In this context, Zielinski et al (2018) and Rad and Barbosa (2012) computed controversy in Wikipedia by mining a variety of metadata from logs, including revisions, edits and changes on Wikipedia discussion pages.…”

Section: Controversy In Online Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Exploiting Open Data to analyze discussion and controversy in online citizen participation

Cantador

Cortés-Cediel

Fernández

2020

Information Processing & Management

View full text Add to dashboard Cite

In this paper we propose a computational approach that applies data mining techniques to analyze the citizen participation recorded in an online digital platform. Differently to previous work, the approach exploits external knowledge extracted from Open Government Data for processing the citizens' proposals and debates of the platform, enabling to characterize targeted issues and problems, and analyze the levels of discussion, support and controversy raised by the proposals. As a result of our analysis, we derive a number of insights and conclusions of interest and value for both citizens and government stakeholders in decision and policy making tasks. Among others, we show that proposals targeting issues that affect large majorities tend to be supported by citizens and ultimately implemented by the city council, but leave aside other very important issues affecting minority groups. Our study reveals that most controversial, likely relevant, problems do not always receive sufficient attention in e-participation. Moreover, it identifies several types of controversy, related to ideological and socioeconomic factors and political attitudes.

show abstract

Section: Controversy In Online Discussionmentioning

confidence: 99%

Section: Controversy In Online Discussionmentioning

confidence: 99%

Exploiting Open Data to analyze discussion and controversy in online citizen participation

Cantador

Cortés-Cediel

Fernández

2020

Information Processing & Management

View full text Add to dashboard Cite

show abstract

“…Bykau et al (2015) described a new algorithm based on user editing history to detect controversy in Wikipedia. An automatic controversy detection model based on a formal model was applied to Wikipedia (Zielinski, Nielek, Wierzbicki, & Jatowt, 2018). Jhandir, Tenvir, On, Lee, and Choi (2017) detected controversy using natural language processing models.…”

Section: Machine-learning-based Quality Assessment Methodsmentioning

confidence: 99%

Assessing the quality of information on wikipedia: A deep‐learning approach

Wang

2019

Asso for Info Science & Tech

View full text Add to dashboard Cite

Currently, web document repositories have been collaboratively created and edited. One of these repositories, Wikipedia, is facing an important problem: assessing the quality of Wikipedia. Existing approaches exploit techniques such as statistical models or machine leaning algorithms to assess Wikipedia article quality. However, existing models do not provide satisfactory results. Furthermore, these models fail to adopt a comprehensive feature framework. In this article, we conduct an extensive survey of previous studies and summarize a comprehensive feature framework, including text statistics, writing style, readability, article structure, network, and editing history. Selected state-of-the-art deep-learning models, including the convolutional neural network (CNN), deep neural network (DNN), long short-term memory (LSTMs) network, CNN-LSTMs, bidirectional LSTMs, and stacked LSTMs, are applied to assess the quality of Wikipedia. A detailed comparison of deep-learning models is conducted with regard to different aspects: classification performance and training performance. We include an importance analysis of different features and feature sets to determine which features or feature sets are most effective in distinguishing Wikipedia article quality. This extensive experiment validates the effectiveness of the proposed model.

show abstract

“…Focusing on the subject of a web page, it can be divided into a classification problem of which category the web page belongs to and a detection problem of which event/action the web page belongs to. In the classification problem, genre classification [17], controversy classification [18], and emotion classification [19] have been proposed. Rumor detection [20], phishing detection [21], fraud detection [22], and fake detection [23] have been proposed in the detection problem.…”

Section: A Web Page Classificationmentioning

confidence: 99%

High-Quality Train Data Generation for Deep Learning-Based Web Page Classification Models

Kim

Lee

2021

IEEE Access

View full text Add to dashboard Cite

The current deep learning models detecting relevant web pages show low accuracy because of the poor quality of the training data. In this paper, we propose a novel algorithm to automatically generate high-quality training data based on the frequency of the document including the entity of interest. Our experimental results with movies and cellphones data sets show that the average F 1 -score of the deep learning models (FNN, CNN, Bi-LSTM, and SeqGAN) trained with our proposed algorithm shows up to 0.9992 in F 1 -score.

show abstract

Computing controversy: Formal model and algorithms for detecting controversy on Wikipedia and in search queries

Cited by 20 publications

References 20 publications

Exploiting Open Data to analyze discussion and controversy in online citizen participation

Exploiting Open Data to analyze discussion and controversy in online citizen participation

Assessing the quality of information on wikipedia: A deep‐learning approach

High-Quality Train Data Generation for Deep Learning-Based Web Page Classification Models

Contact Info

Product

Resources

About