Automatic Extraction of Property Norm‐Like Data From Large Text Corpora

Traditional methods for deriving property‐based representations of concepts from text have focused on either extracting only a subset of possible relation types, such as hyponymy/hypernymy (e.g., car is‐a vehicle ) or meronymy/metonymy (e.g., car has wheels ), or unspecified relations (e.g., car — petrol ). We propose a system for the challenging task of automatic, large‐scale acquisition of unconstrained, human‐like property norms from large text corpora, and discuss the theoretical implications of such a system. We employ syntactic, semantic, and encyclopedic information to guide our extraction, yielding concept‐relation‐feature triples (e.g., car be fast , car require petrol , car cause pollution ), which approximate property‐based conceptual representations. Our novel method extracts candidate triples from parsed corpora (Wikipedia and the British National Corpus) using syntactically and grammatically motivated rules, then reweights triples with a linear combination of their frequency and four statistical metrics. We assess our system output in three ways: lexical comparison with norms derived from human‐generated property norm data, direct evaluation by four human judges, and a semantic distance comparison with both WordNet similarity data and human‐judged concept similarity ratings. Our system offers a viable and performant method of plausible triple extraction: Our lexical comparison shows comparable performance to the current state‐of‐the‐art, while subsequent evaluations exhibit the human‐like character of our generated properties.     

Anger in Wikipedia: Perceived intentionality and threat appraisal as mediators of anger about man‐made attacks

The online encyclopedia Wikipedia has established strict guidelines for the objectivity of content. At the same time, Wikipedia includes articles on negative events, such as disasters or man‐made attacks. These events can elicit strong emotions, which in turn may spill over into Wikipedia articles. Previous research has shown that Wikipedia articles on man‐made attacks contain more anger‐related content than Wikipedia articles on disasters. Building on these findings, we aimed to investigate whether the threat that Wikipedia authors experience when they learn of an attack is relevant as a factor in explaining the anger effect. Threat is known to elicit active and engaged reactions, such as anger, which is why it is a likely explaining factor. Our research also aimed to replicate the findings from the linguistic analysis of the Wikipedia articles using controlled scenario‐based laboratory experiments. Three studies demonstrated that man‐made attacks (terrorist attack, shooting rampage) elicited more threat, more anger, and more expressions of anger‐related content in Wikipedia texts than nature‐made disasters (earthquake, flood) and man‐made disasters (train accident). Moreover, threat was relevant as a factor in explaining the effects on anger and anger‐related content in the Wikipedia texts. These mediations could be explained by the perceived intentionality of the event. This research highlights the findings that perceived intentionality and threat are relevant mediating factors for feelings and expressions of anger after man‐made attacks.