Exploring the grouning of linguistic concepts in images on the Web

点击翻译按钮获取中文摘要

The recent research direction toward multimodal semantic representation would be further advanced, if we could have a machinery to collect adequate images from the Web, given a target concept. With this motivation, this research particularly investigated into the Web imageabilities of the behavioral features of a basic-level concept. The primary contributions made in this research are twofold: (1) "beaver building dams"-type queries can better yield relevant Web images, suggesting that the present participle form ("-ing" form) of a verb, as a query component, is more effective than the base form; (2) the behaviors taken by animate beings are likely to be more depicted on the Web, particularly if the behaviors are, in a sense, inherent to animate beings (e.g., motion, consumption). Web further analyzed linguistic annotations that were independently given to some of the images, and discusses an aspect of the semantic gap between image and language.