Semantic web technologies for the information retrieval on wikidata

Abstract

Information Retrieval is responsible for the storage and automatic retrieval of information, and these documents may consist of texts, web pages, audio, video, images, graphics and figures. Information Retrieval techniques have gained importance with the growth of the Web, because the unlimited amount of information can express the most diverse forms and levels of quality to what is expected. With this in mind, the present work studies methods and technologies capable of retrieving this information, focusing on searching structured databases called Linked Data, but specifically on the Wikidata project, a database structured using Semantic Web concepts, which brings together the knowledge from Wikipedia. Seeking to understand how this information retrieval is done in the Wikidata project, this research has the objective of presenting the media that Wikidata provides to RI and how they use the principles of the Semantic Web. The methodology used was an exploratory study based on the research and applied, since tests were done in the database of Wikidata. As a result, the characteristics of the various forms of data access and retrieval were identified, tracing the correlations between each of these forms, with the theoretical framework of the Semantic Web and Information Retrieval. It was concluded that Wikidata stands as a solid database, with a large volume of contents, quite current, that has a series of recovery mechanisms, capable of serving the most diverse applications on the Web, because these mechanisms are built with different technologies and configurations.