Difference between revisions of "History-Based Article Quality Assessment on Wikipedia"
(Overview - History-Based Article Quality Assessment on Wikipedia) |
(wikilinks) |
||
| Line 1: | Line 1: | ||
| − | '''History-Based Article Quality Assessment on Wikipedia''' - scientific work related to Wikipedia quality published in 2018, written by Shiyue Zhang, Zheng Hu, Chunhong Zhang and Ke Yu. | + | '''History-Based Article Quality Assessment on Wikipedia''' - scientific work related to [[Wikipedia quality]] published in 2018, written by [[Shiyue Zhang]], [[Zheng Hu]], [[Chunhong Zhang]] and [[Ke Yu]]. |
== Overview == | == Overview == | ||
| − | Wikipedia is widely considered as the biggest encyclopedia on Internet. Quality assessment of articles on Wikipedia has been studied for years. Conventional methods addressed this task by feature engineering and statistical machine learning algorithms. However, manually defined features are difficult to represent the long edit history of an article. Recently, researchers proposed an end-to-end neural model which used a Recurrent Neural Network(RNN) to learn the representation automatically. Although RNN showed its power in modeling edit history, the end-to-end method is time and resource consuming. In this paper, authors propose a new history-based method to represent an article. Authors also take advantage of an RNN to handle the long edit history, but authors do not abandon feature engineering. Authors still represent each revision of an article by manually defined features. This combination of deep neural model and feature engineering enables model to be both simple and effective. Experiments demonstrate model has better or comparable performance than previous works, and has the potential to work as a real-time service. Plus, authors extend model to do quality prediction. | + | Wikipedia is widely considered as the biggest encyclopedia on Internet. Quality assessment of articles on [[Wikipedia]] has been studied for years. Conventional methods addressed this task by feature engineering and statistical machine learning algorithms. However, manually defined [[features]] are difficult to represent the long edit history of an article. Recently, researchers proposed an end-to-end neural model which used a Recurrent Neural Network(RNN) to learn the representation automatically. Although RNN showed its power in modeling edit history, the end-to-end method is time and resource consuming. In this paper, authors propose a new history-based method to represent an article. Authors also take advantage of an RNN to handle the long edit history, but authors do not abandon feature engineering. Authors still represent each revision of an article by manually defined features. This combination of deep neural model and feature engineering enables model to be both simple and effective. Experiments demonstrate model has better or comparable performance than previous works, and has the potential to work as a real-time service. Plus, authors extend model to do quality prediction. |
Revision as of 11:30, 3 December 2020
History-Based Article Quality Assessment on Wikipedia - scientific work related to Wikipedia quality published in 2018, written by Shiyue Zhang, Zheng Hu, Chunhong Zhang and Ke Yu.
Overview
Wikipedia is widely considered as the biggest encyclopedia on Internet. Quality assessment of articles on Wikipedia has been studied for years. Conventional methods addressed this task by feature engineering and statistical machine learning algorithms. However, manually defined features are difficult to represent the long edit history of an article. Recently, researchers proposed an end-to-end neural model which used a Recurrent Neural Network(RNN) to learn the representation automatically. Although RNN showed its power in modeling edit history, the end-to-end method is time and resource consuming. In this paper, authors propose a new history-based method to represent an article. Authors also take advantage of an RNN to handle the long edit history, but authors do not abandon feature engineering. Authors still represent each revision of an article by manually defined features. This combination of deep neural model and feature engineering enables model to be both simple and effective. Experiments demonstrate model has better or comparable performance than previous works, and has the potential to work as a real-time service. Plus, authors extend model to do quality prediction.
