As reliance on electronic resources increases for business, government, research, and education, it is realized that numerous applications could be improved by language understanding capabilities that go beyond those which are currently available. The Center for Natural Language Processing (CNLP) has accepted this challenge and is focused on developing a breadth of NLP technologies that build on its core capability to recognize and represent both the explicit and implicit content of texts. In this talk, we’ll discuss a number of these widely applicable capabilities, and then talk specifically about applica-tions that CNLP has developed for NSF’s National Science Digital Library (NSDL) Program.
The talk will also include a demo of the Content Assignment Tool (CAT) that applies NLP to learning resources, such as lesson plans, to automatically suggest relevant standards for each resource. The user selects the appropriate standards from the suggest-ions to be added as metadata. By applying machine-learning, CAT learns from the vetted assignments and improves future suggestions. CAT’s functionalities are useful for collection builders, catalogers, curriculum developers, or individual teachers who have valuable resources to contribute to the STEM NSDL.
This invited presentation is part of the "NSDL at NSF" Luncheon Lecture Series
