Towards Computer Understanding of Direct Object Defaults

The paper addresses the phenomenon of direct object defaults in text as part of exploring the meaning of the unsaid and making it accessible for computer understanding. It describes a large but reasonably simple computer experiment on the basis of one hypothesis about defaults, namely, that a true default may appear as a direct object only if modified, e.g., Bob ate fresh food but? Bob ate food. The algorithm reduced over 24,000 occurrences, in Wikipedia for Schools, of the 200 most frequent verbs with and without direct objects, and those direct objects with and without modifiers, to over 200,000 default candidates. A discussion dealt with ways of restricting this list further to actual defaults.

I. Introduction

With expanding computational abilities and technological advances, increasingly more emphasis is being put on information that can be received from text automatically. Moreover, by combining semantic resources with advanced statistical processing, machines are now becoming more and more successful in processing such information and creating proper links between entities to come up with answers that sometimes surpass human knowledge. What still remains to be discovered, however, is whether such systems are capable of understanding things that are so obvious to humans that we omit to verbalize them or to write them down.