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

Abstract

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.

A deep learning-based quality assessment model of collaboratively edited documents: A case study of Wikipedia

“…Contrary to what we expect, the CNN performs the worst. In most cases, the CNN has high performance in learning relevant features and ruling out irrelevant features [54,62]. Moreover, after comparison of basic LSTM and CNN-LSTM, we find that the CNN degrades the model performance.…”

“…Few studies have analysed and summarised the existing work. In this section, we perform an extensive review of the existing feature frameworks [1,2,5,6,12,23–27,35,42,44,47–54] and propose a comprehensive feature framework as a representation of Wikipedia articles. Text statistics are indicators that measure basic article statistics [1,23], including word count and character count.…”

“…Usually, more editing will make Wikipedia articles more understandable and readable. Network features are based on a Wikipedia link graph [1,23,54]. Network features focus on relationships among different Wikipedia articles.…”

Wikipedia citations: A comprehensive data set of citations with identifiers extracted from English Wikipedia

“…Importantly, such mapping should consider disciplinary differences in citations from Wikipedia, as well as books (5.3 million citations by our estimates) and nonscientific sources such as news outlets and other online media (21.5 million citations), which make up the largest share of Wikipedia citations. Answering these questions is critical to inform the community work on improving Wikipedia by finding and filling knowledge gaps and biases, at the same time guaranteeing the quality and diversity of the sources Wikipedia relies upon (Hube, 2017;Mesgari et al, 2015;Piscopo, Kaffee et al, 2017;Piscopo & Simperl, 2019;Wang & Li, 2020). Link prediction in general, and citation recommendation in particular, have been explored for Wikipedia for some time (Fetahu et al, 2016;Paranjape, West et al, 2016;Wulczyn, West et al, 2016).…”

Measuring Quality of Wikipedia Articles by Feature Fusion‐based Stack Learning

“…To solve that problem, machine learning algorithms have been applied to Wikipedia article quality assessment. They combined machine learning models with hand‐crafted features to assess the quality of Wikipedia articles (Shen et al, 2017; Zhang et al, 2018; Ferschke et al, 2012; Khairova et al, 2017; Wang & Li, 2020; Wang et al, 2019). However, semantic features from the article content are often ignored in these approaches.…”