Autonomous Web-scale Information Extraction

Search engines are extremely useful tools for answering simple questions. However, for more complex questions -- e.g., "which nanotechnology companies are hiring on the West Coast?" -- existing search engines are less effective, because the answers are not contained on just a single page. Answering these questions requires extracting and synthesizing information across multiple documents. Currently, this is a tedious and error-prone manual process. In this talk, Dr. Downey will describe his research aimed at automating the extraction of information from the Web. He will present a model of the redundancy inherent in the Web, and show that the model can be used to identify correct extractions autonomously, without the manually labeled examples typically assumed in previous information extraction researc h. Further, while the redundancy-based model alone is ineffective for the "long tail" of infrequently mentioned facts, Dr. Downey will illustrate how unsupervised language models can be leveraged to overcome this limitation.