A Hybrid Method for Detecting Outdated Information in Wikipedia Infoboxes

A Hybrid Method for Detecting Outdated Information in Wikipedia Infoboxes - scientific work related to Wikipedia quality published in 2013, written by Thong Tran and Tru H. Cao.

Overview

Wikipedia has grown fast and become a major information resource for users as well as for many knowledge bases derived from it. However it is still edited manually while the world is changing rapidly. In this paper, authors propose a method to detect outdated attribute values in Wikipedia infoboxes by using facts extracted from the general Web. Authors proposed method extracts new information by combining pattern-based approach with entity-search-based approach to deal with the diversity of natural language presentation forms of facts on the Web. Authors experimental results show that the achieved accuracies of the proposed method are 70% and 82% respectively on the chief-executive-officer attribute and the number-of-employees attribute in company infoboxes. It significantly improves the accuracy of the single pattern-based or entity-search-based method. The results also reveal the striking truth about the outdated status of Wikipedia.