Wikidata, A Semantic Database

by Chris S
In this article, Wikimedia recently announced a new project of theirs called Wikidata. This is suppose to be a new database of information that can be read and edited but people and machines. One of the main goals of Wikidata is the development of an actual semantic, machine-readable database. The German Chapter of Wikimedia is handling the initial project with an estimated completion date of March 2013. There are three phases in the development to Wikidata. The first one is basically creating a Wikidata page for every single Wikipedia entry spanning over Wikipedia’s 280 supported languages. Phase two consists of user editors being able to use, and add data to the project. The last phase is suppose to allow for automatic creation of charts and lists going off of the data in Wikidata. This can then be populated into Wikipedia pages. If all goes according to plan, the new database of knowledge readable by people and machines alike will be created.

One thing I found interested that Wikidata would have a feature that will enable users to ask questions and will be answered automatically. The idea behind it is to replace the queries we might find using MySQL, the manually created structured answers. Also, Google mentioned that their efforts were increased when they tried to provide direct answers to queries. Common questions that are searched would be answered with a direct answer, and Wikidata can greatly help that.

I for one would love to have an answer for some of the things I search for. If Wikidata can make an improvement with that then I am all for it and cannot wait for it to be completed. The idea of a semantic database is intriguing. If this does work out, as stated in the article, we may see more actual answers to our queries rather than links to Wikipedia pages with the information we need.

 

Perez, Sarah. “Wikipedia€™s Next Big Thing: Wikidata, A Machine-Readable, User-Editable Database Funded By Google, Paul Allen And Others.” TechCrunch. 30 Mar. 2012. Web. 12 Apr. 2012. <http://techcrunch.com/2012/03/30/wikipedias-next-big-thing-wikidata-a-machine-readable-user-editable-database-funded-by-google-paul-allen-and-others/>.

3 thoughts on “Wikidata, A Semantic Database”

  1. This definately seems like an interesting route to go for wikipedia, considering how much information it provides already. While this seems like the next step to making information more readily available to people, I feel like this feature is relatively unnecessary at the moment. I can only imagine how rudimentary the answering system will be, providing only answers to easy questions or questions that have been worded out in a perfect format, which will stay make Googling the question the better choice. However, if this technology is simply a step towards a much more expansive goal of creating a complete all-knowing guru system that answers any question no matter how casually worded, this could be heading in the right direction. But this goal is similar to the lofty idea of creating a system that can understand Re-Captcha phrases, which is near impossible as it is.

  2. In response to the other comment (by Michael), I believe what Wikidata really has to offer is not simply the answer, but a more complex, thorough answer system that includes related topics, visual data, and for extremely numeric or statistical questions, Wikidata can develop a more reliable and thorough answer than just Googling it could. For instance, a user could ask Wikidata, “which top five cities in the United States have the most steel production?” and Wikidata would utilize all information to develop a coherent answer.

  3. I love this idea of a semantic database. Already, Google has primitive semantic kind of searches where they automatically find an answer to your question “capital of X city,” it’d be amazing to have this technology advanced to answer more complex questions.

Comments are closed.