Open and image-open relations

Identifying Relations for Open Information Extraction

Open Information Extraction (IE) is the task of extracting assertions from massive corpora without requiring a pre-specified vocabulary. This paper shows that the output of state-ofthe-art Open IE systems is rife with uninformative and incoherent extractions. To overcome these problems, we introduce two simple syntactic and lexical constraints on binary relations expressed by verbs. We implemen...

Open Problems with Part-Whole Relations

Open Problems with Part-Whole Relations

Among attributes, a fundamental role is often played by various forms of part-whole relations, which contribute to describe the structure of composite objects. The representation of such a structural information usually requires a particular semantics together with specialized inference, but rarely do current modeling formalisms and methodologies give it a specific, "first-class" dignity [Simon...

Open standards, open formats, and open source

The paper proposes some comments and reflections on the notion of “openness” and on how it relates to three important topics: open standards, open formats, and open source. Often, these terms are considered equivalent and/or mutually implicated: “open source is the only way to enforce and exploit open standards”. This position is misleading, as it increases the confusion about this complex and ...

Taxonomic and Ecological Relations in Open-ended Induction

Recent research suggests that inductive inference may be based on salient relations between premise and/or conclusion categories (Medin, Coley, Storms and Hayes, 2003). The types of relations deemed relevant may be influenced by experience (e.g. López, Atran, Coley, Medin and Smith, 1997; Shafto & Coley, 2003). Novices in a given domain have been shown to rely primarily on taxonomic relationshi...