Text Analytics

Catalog description

This course will first cover techniques for compiling textual corpora from web pages, pdfs, scanned pdfs, images, audio clips, etc. Secondly, it will look at processes for extracting some common types of information from these corpora. In particular, we will cover extracting named entities (persons, locations, organizations, etc.), relations between entities, events, transactions, topics, document summaries, abstracts, legal clauses, etc. This course is different from standard courses in Natural Language Processing and Computational Linguistics in that we will spend significant amount of course time on compiling textual corpora from documents in a variety of formats and our emphasis will be on extracting information that can be fed to analytics pipelines.