Granularity in Natural Language Discourse

This paper discusses the phenomenon of granularity in natural language. By ‘granularity’ we mean the level of detail of description of an event or object. Humans can seamlessly shift their granularity perspective while reading or understanding a text. To emulate this mechanism, we describe a set of features that identify the levels of granularity in text, and empirically verify this feature set using a human annotation study for granularity identification. This theory is the foundation for any system that can learn the (global) behavior of event descriptions from (local) behavior descriptions. This is the first research initiative, to our knowledge, for identifying granularity shifts in natural language descriptions.