Wikipedia infoboxes summarize the most important information of a Wikipedia page. In addition to being useful for Wikipedia readers, knowledge bases like DBpedia and Google Knowledge Graph can easily process and use their semi-structured content. However, the current mechanisms for creating or extending infoboxes are difficult and complex for editors to use. The result is fewer Wikipedia pages with infoboxes and more errors, inaccuracies, and deficiencies in existing infoboxes.
We present an approach to help users create rich and accurate Wikipedia infoboxes. Our system uses statistical and semantic knowledge from linked data sources to simplify the process of creating and extending infobox content. It constructs semantic infobox templates by suggesting common attributes and their values from similar articles and controlling the expected values semantically. We present the implemented prototype and experiments with users that evaluate both the quality of the new infobox content and the users' subjective experience.
Useful links and references
Roberto Yus
Assistant Professor
My research interests include Data Management, Knowledge Representation, the Internet of Things, and Privacy.
