Commonsense LocatedNear Relation Extraction

Artificial Intelligent systems can benefit from incorporating commonsense knowledge as background, such as ice is cold (HASPROPERTY), chewing is a sub-event of eating (HASSUBEVENT), chair and table are typically found near each other (LOCATEDNEAR), etc. This kind of commonsense facts have been utilized in many downstream tasks, such as textual entailment [4, 1] and visual recognition tasks [29]. The commonsense knowledge is often represented as relation triples in commonsense knowledge bases, such as ConceptNet by MIT [20], one of the largest commonsense knowledge graph available today. However, this kind of commonsense knowledge bases are usually manually curated or crowd-sourced by community efforts and thus do not scale well.