Ultimately we would like our machines to not only search and retrieve information, but also have some "understanding" of the material that they are manipulating so that they can better meet the user's needs. In this talk, I will present our work in Project Halo to create an (iPad hosted) "'knowledgeable biology textbook", called Inquire. Inquire includes a formal, hand-crafted knowledge base encoding some of the book's content, being augmented (this year) with capabilities for textual entailment and question-answering directly from the book text itself. Inquire allows the user to not only read and browse the textbook, but also to ask questions and get reasoned or retrieved answers back, explore the material through semantic connections, and receive suggestions of useful questions to ask. In this talk I will describe the project, in particular the textual question-answering component and its use of natural language processing, paraphrasing, textual entailment, and its exploitation of the formal knowledge base. I will also discuss the interplay being developed between the hand-built knowledge and automatic text-extracted knowledge, how each offers complementary strengths, and how each can leverage the other. Finally I will discuss the value of this approach, and argue for the importance of creating a deeper understanding of textual material, and ultimately more knowledgeable machines.
展开▼