Mining Possessions: Existence, Type and Temporal Anchors

Proceedings of NAACL-HLT 2018, pages 496–505 New Orleans, Louisiana, June 1 - 6, 2018. c⃝2018 Association for Computational Linguistics Mining Possessions: Existence, Type and Temporal Anchors Dhivya Chinnappa and Eduardo Blanco Human Intelligence and Language Technologies Lab University of North Texas Denton, TX, 76203 dhivyainfantchinnappa@my.unt.edu, eduardo.blanco@unt.edu Abstract This paper presents a corpus and experiments to mine possession relations from text. Specif- ically, we target alienable and control posses- sions, and assign temporal anchors indicating when the possession holds between possessor and possessee. We present new annotations for this task, and experimental results using both traditional classifiers and neural networks. Re- sults show that the three subtasks (predicting possession existence, possession type and tem- poral anchors) can be automated. 1 Introduction Every language has a way of expressing posses- sive relationships (Aikhenvald and Dixon, 2012). Possession is an asymmetric semantic relation be- tween two entities, where one entity (the pos- sessee) belongs to the other entity (the possessor) (Stassen, 2009). When it comes to defining pos- session, belongs includes a wide range of relation- ships, including (hereafter, we use x to refer to the possessor, and y to refer to the possessee) kin- ship (e.g., [my]x oldest [son]y), part-whole (e.g., the [car]x’s [dashboard]y), physical and tempo- rary possession (e.g., [I]x have John’s [book]y), possession of something intangible (e.g., [John]x got the [flu]y last year) and proximity (e.g., The [shelf]x has a [glass sculpture]y). Possession relations can be divided into alien- able (also referred to as acquired, transferable, non-intimate, etc.) and inalienable (also referred to as inherent, inseparable, intimate, etc.). Pos- sessees that can be separated from their posses- sors are alienable, and possessees that cannot nor- mally be separated from their possessors are in- alienable (Heine, 1997). For example, [John]x’s [condo]y is alienable, and [John]x’s [arm]y is in- alienable (some previous works would call the lat- ter a part-whole relation instead). Tham (2004) defines control possession as a relation in which the possessor has temporary control of the pos- sessee, but does not necessarily alienably possess it (e.g., [John]x borrowed the [car]y for the week- end). Following the aforecited works, possession goes beyond ownership of property. Possession relations can be expressed in a wide variety of syntactic constructions, including noun phrases (e.g., [John]x’s [car]y) and clauses (e.g., [John]x bought a [blue car]y). The subject of a verb can map to either the possessor as exempli- fied above, or to the possessee (e.g., The [car]y be- longs to [John]x) (Aikhenvald and Dixon, 2012). Within computational linguistics, possession re- lationships have usually been studied as part of larger studies that target all relations between ar- guments connected with a syntactic pattern (e.g., possessive constructions, nominals). Addition- ally, previous efforts have mostly targeted alien- able possession—or alternatively, ownership. The work presented here takes a different approach. We start by pairing people (plausible possessors) with physical objects (plausible possesses). Then, we determine whether a possession relationship exists, and if so, (a) determine the type (alienable or control) and (b) assign temporal anchors with respect to …

Content truncated to 500 words