The Semantic Web: there and back again

The Web has made us all smarter by providing ubiquitous and easy access to virtually all human knowledge. Most of the knowledge is encoded as text, images or videos -- representations that people can easily process, understand and use but which still are difficult for machines. Making this knowledge more accessible to machines will unleash new and more intelligent computer applications and systems. Enabling computers to fully understand text, images and videos is a long term goal that will require many decades to fully achieve. An alternative is to represent use Web-based technologies to directly represent information and knowledge, tying the data to semantics models, facilitating data sharing and linking, and supporting provenance annotations. This is the value proposition offered by the Tim Berners-Lee's Web of Data vision that is being adopted by search engines, Wikipedia and advanced applications.

While the concepts and technologies underlying the Semantic Web are mature and supported by sound standards, their use within applications remains relatively low. This talk will discuss current research aimed at reducing barriers to wider adoption and use. We will describe approaches to extracting semantic data from text, publishing and sharing scientific data on social media systems, generate Semantic Web data from tables and spreadsheets, and new approaches to querying large collections of Semantic Web data.