Text and Meaning: Understanding Content

What can and should your machine understand about what you are saying? On a typical day, you may browse several web pages, read and write emails, update your Facebook status, follow your favorite Twitterers, exchange text messages with friends, and so forth. All of these communications are machine-readable—and therefore, potentially machine-understandable.

We are entering an era in which mainstream users of technology products expect the systems they interact with to handle text in an intelligent manner. This session will explore the power and limitations of contemporary production-ready and research-grade text processing technologies which are being developed to meet these expectations. We will cover a variety of applications of semantic analysis and language processing, and discuss when and how these are best used—and the surprising interactions between user experience design choices and the success of any such AI system.

As a case study, we will focus on the diverse range of technologies used to build Aardvark, a social search engine which combines advanced text processing techniques with human intelligence to find answers to questions.