{ "paper_id": "2021", "header": { "generated_with": "S2ORC 1.0.0", "date_generated": "2023-01-19T07:12:18.192731Z" }, "title": "The ISA-17 Quantification Challenge: Background and introduction", "authors": [ { "first": "Harry", "middle": [], "last": "Bunt", "suffix": "", "affiliation": { "laboratory": "", "institution": "Tilburg University", "location": { "settlement": "Tilburg", "country": "Netherlands" } }, "email": "bunt@tilburguniversity.edu" } ], "year": "", "venue": null, "identifiers": {}, "abstract": "This short paper provides background information for the shared quantification annotation task at the ISA-17 workshop, a.k.a. the Quantification Challenge. The role of the abstract and concrete syntax of the QuantML markup language are explained, and the semantic interpretation of QuantML annotations in relation to the ISO principles of semantic annotation. Additionally, the choice of the test suite of the Quantification Challenge is motivated.", "pdf_parse": { "paper_id": "2021", "_pdf_hash": "", "abstract": [ { "text": "This short paper provides background information for the shared quantification annotation task at the ISA-17 workshop, a.k.a. the Quantification Challenge. The role of the abstract and concrete syntax of the QuantML markup language are explained, and the semantic interpretation of QuantML annotations in relation to the ISO principles of semantic annotation. Additionally, the choice of the test suite of the Quantification Challenge is motivated.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Abstract", "sec_num": null } ], "body_text": [ { "text": "The ISA-17 Quantification Challenge was motivated by the decision of the International Organisation for Standardisation ISO to develop an international standard for the annotation of quantification in natural language, extending the series of standards for semantic annotation called the ISO Semantic Annotation Framework (SemAF, ISO 24617)). Other parts of this series include standards for the annotation of (1) time and events (ISO 24617-1, 'ISO-TimeML');", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(2) dialogue acts (ISO 24617-2, 'DiAML');", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(3) semantic roles (ISO 24617-4);", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(4) spatial information (ISO 24617-7, 'ISO-Space');", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(5) discourse relations (ISO 24617-8, 'DR-Core');", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "(6) coreference (ISO 24617-9, 'Reference Annotation Framework' ).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "Also belonging to this series is the meta-standard ISO 24617-6, 'Principles of semantic annotation', which defines a common methodological framework for developing other parts of SemAF.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "As the first steps in the development of an annotation standard for quantification, preliminary studies have been conducted and reported at LREC 2018 (Bunt, Lee and Pustejovsky, 2018) , at IWCS 2019 (Bunt, 2019) , and in a technical report of Tilburg University (Bunt, 2021) in which the markup language QuantML is defined. On the basis of these studies, the document ISO WD 24617-12 was drafted. The ISA-17 Quantification Challenge is intended to identify the strengths, limitations, and deficiencies of the QuantML proposal by inviting experts in quantification and/or in semantic annotation to explore the application of QuantML in the annotation of a range of test sentences that display some of the phenomena that the future standard would hope to cover.", "cite_spans": [ { "start": 150, "end": 183, "text": "(Bunt, Lee and Pustejovsky, 2018)", "ref_id": "BIBREF10" }, { "start": 199, "end": 211, "text": "(Bunt, 2019)", "ref_id": null }, { "start": 262, "end": 274, "text": "(Bunt, 2021)", "ref_id": "BIBREF9" } ], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "This paper is organised as follows. Section 2 briefly summarizes the ISO Principles of semantic annotation, especially where it concerns the architecture of a semantic annotation scheme, including the design of an abstract syntax and the specification of a concrete syntax plus the significance of a compositional semantics of the (abstract syntactic structures of the) annotations, and applies this to the QuantML annotation scheme. Section 3 introduces and motivates the choices in the test suite used in the Quantification Challenge.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1" }, { "text": "The usual definition of a markup language consists of the specification of a number of XML elements, attributes, and values, that can be used to form descriptions of the linguistic properties of certain stretches of text or speech, called 'markables'. The definitions of TimeML (Pustejovsky et al., 2007) and SpatialML (Mani et al, 2010) illustrate this.", "cite_spans": [ { "start": 278, "end": 304, "text": "(Pustejovsky et al., 2007)", "ref_id": "BIBREF25" }, { "start": 319, "end": 337, "text": "(Mani et al, 2010)", "ref_id": "BIBREF21" } ], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "According to the ISO Principles of semantic annotation ISO 24617-6; see also Bunt (2015) and Pustejovsky et al. (2017) ) a semantic annotation scheme has a three-part architecture consisting of (1) an abstract syntax that specifies the possible annotation structures at a conceptual level as settheoretical constructs, such as pairs and triples of concepts; (2) a concrete syntax, that specifies a representation format for annotation structures (for example using XML); (3) a semantics that specifies the meaning of the annotation structures defined by the abstract syntax.", "cite_spans": [ { "start": 77, "end": 88, "text": "Bunt (2015)", "ref_id": "BIBREF5" }, { "start": 93, "end": 118, "text": "Pustejovsky et al. (2017)", "ref_id": "BIBREF24" } ], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "The distinction of an abstract and a concrete syntax is motivated by the fundamental distinction between 'annotations' and 'representations', made in the Linguistic Annotation Framework (ISO standard 24612, see also Ide and Romary (2004) . An 'annotation' captures certain linguistic information, independent on a particular representation format, while a 'representation' specifies a format for representing annotations. In the three-part architecture, 'annotations' are the conceptual structures defined by the abstract syntax (and called 'annotation structures'); 'representations' correspond to the particular format in which these structures are expressed (which we will usually call 'annotations', following the most common usage of this term). ISO standards for semantic annotation are intended to apply not at the level of representation formats, but that of the information they represent: the level of conceptual annotation structures.", "cite_spans": [ { "start": 216, "end": 237, "text": "Ide and Romary (2004)", "ref_id": "BIBREF14" } ], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "The third component of a semantic annotation scheme, the specification of a semantics of annotation structures, is a requirement specific of semantic annotations, the requirement of semantic adequacy (Bunt and Romary (2002) ): if the annotations would not have a well-defined semantics, it would not be clear what semantic information they add to the natural language expressions they annotate. Defining the semantics at the level of the abstract syntax puts the focus of an annotation standard at the conceptual level, rather than at the level of representation formats.", "cite_spans": [ { "start": 200, "end": 223, "text": "(Bunt and Romary (2002)", "ref_id": "BIBREF11" } ], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "Formally, the definition of an annotation scheme is a triple consisting of specifications of an abstract syntax (AS), a concrete syntax (CS), and a semantics (ASem):", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "(1) A = ASyn a , ASyn c , ASem", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "The abstract syntax consists of the specification of a set of basic concepts, called the 'conceptual inventory' (CI), and a set of constructions (AC) for forming conceptual structures out of basic concepts.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "(2) ASyn a = CI, AC Together, the sets CI and AC define the class of well-formed annotation structures.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "The concrete syntax specification ASyn C contains a vocabulary V c , the specification CC of a class of syntactic structures, such as XML elements, and an encoding function F e . The components V c and CC together define a class of well-formed representations, and F e assigns such a representation to every well-formed annotation structure.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "(3)", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "ASyn c = V c , CC, F e", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "The semantics ASem can be specified in various ways, for example as a model-theoretic semantics M, I M with a model M and an interpretation function I M that assigns concepts from M as meanings to annotation structures.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "The three parts of the annotation schema are related through the encoding function F e and the interpretation function I M . In particular, a requirement for the relation between abstract and concrete syntax is that the concrete syntax is complete and unambiguous (Bunt (2010)) for the abstract syntax, i.e. every annotation structure has a representation in the concrete syntax, and every representation is the encoding of exactly one annotation structure. In other words, F e is a total function and so is its inverse F \u22121 e . The semantic component should also be complete: every annotation structure has a semantic interpretation.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "Two types of structure are distinguished in an abstract syntax: entity structures and link structures. An entity structure contains semantic information about a segment of primary data and is formally a pair m, s consisting of a markable (m), which refers to a segment of primary data, and certain semantic information (s). A link structure contains information about the way two or more segments of primary data are semantically related. In QuantML three types of entity structure are defined (participant structures, event structures, and modifier structures) and two types of link structure (participation links and scope links). Participation links relate participants to events; scope links indicate scope relations between participants. See further Section 2.3.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "The three-part structure of a semantic annotation scheme does not need to frighten the users of such a scheme: annotators (human or automatic) only have to deal with concrete representations. They can rely, however, on the abstract syntax and its semantics that comes with the definition of the scheme, in particular when in doubt how to use the concrete syntax for annotating certain linguistic phenomena: rather than just relying on annotation guidelines, which are bound to be incomplete, they check the semantics for the precise implications of the choices offered by the concrete syntax.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Annotation scheme architecture", "sec_num": "2.1" }, { "text": "A concrete QuantML syntax is specified here in the form of an XML representation of annotation structures. For each type of entity structure, defined by the abstract syntax, a corresponding XML element is defined; each of these elements has an attribute @xml:id whose value is a unique identification of (the information in) the element, and an attribute @target, whose value anchors the annotation in the primary data, having a markable as value (or a sequence of markables). In addition, these elements have the following attributes:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "1. the XML element , for representing participant structures, has the attributes @domain, @involvement, @definiteness and optionally @size (default value: \u2265 1;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "2. the XML element , for representing event structures, has the attribute @pred for specifying an event type;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "3. the XML element , for representing a quantification domain: has the attributes @source (with multiple values in the case of a conjunctive specification) and @restrictions;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "4. the XML element , for representing quantification source domain specifications without modifiers: has the attributes @pred and @individuation;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "5. the XML element , for representing adjectival modifiers, with the attributes @pred and @distr, and optionally the attribute @restrictions;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "6. the XML element , for representing nouns as modifiers, with the attributes @pred and @distr, and optionally @restrictions;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "7. the XML element , for representing PP modifiers, with the attributes @pRel, @pEntity, @distr and @linking;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "8. the XML element , for representing relative clauses, with the attributes @sem-Role, @clause, @distr and @linking;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "9. the XML element , for representing possessive restrictions, with the attributes @possessor, @distr, and @linking.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "For the two types of link structure defined by the abstract syntax, a corresponding XML element is defined:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "\u2022 has the attributes @event, @participant, @semRole, @distr, and @evScope (default value: \"narrow\"), and optionally @exhaustiveness (default value: \"nonexhaustive\"), @rep (repetitiveness, default value: \u2265 1), and @polarity (default value: \"positive\");", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "\u2022 has the attributes @arg1, @arg2, @scopeRel.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML concrete syntax", "sec_num": "2.2" }, { "text": "The QuantML abstract syntax defines the following entity structures m, s with markable m and semantic content s:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "1. Participant structures: s is a triple or quadruple DS, q, d, N , where DS is a domain specification, q is a specification of domain involvement, d is a definiteness, and N is a numerical size specification (optional).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "2. Event structures: s is a predicate denoting an event domain.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "3. Modifier structures: s contains a predicate for (NP head) noun modification by an adjectives, noun, prepositional phrase, relative clause, or possessive restriction, plus parameters for specifying properties of the modification.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "The following link structures are defined:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "1. Participation links: A 6-9 tuple as shown in (4), where the first two components are the linked event and participant structures, and the other components indicate properties of the way in which the participants are involved in the events, specifying a semantic role (R), a distribution (d), an event scope (\u03c3) that specifies whether the event structure has wider or narrower scope than the participant structure, and optionally an exhaustiveness (\u03be), a repetitiveness (\u03c1), and a polarity (p).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "(4) L P 1 = e , p , R, d, \u03c3, p 2. Scope relation links: triples participation link, participation link, scope relation .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "The conceptual inventory of the abstract syntax includes: 1 1. predicates that characterise quantification domains, corresponding to the meanings of common nouns of the language of the primary data;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "2. predicates that characterise event domains, corresponding to the meanings of verbs (and some other lexical items);", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "3. predicates corresponding to the meanings of adjectives or prepositions;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "4. relations that denote semantic roles; for this purpose, the semantic roles defined in ISO 245617-4 (Semantic roles) are used;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "5. binary and ternary relations for specifying proportional domain involvement, such as most, 'half, 'total', and \"between\";", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "6. non-numerical quantitative predicates for specifying domain involvement, like some and several;", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "7. parameters for specifying definiteness, polarity, distributivity, individuation, relative scoping, repetitiveness and exhaustivity.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "Quantification annotation is associated with the units that in linguistics are called (small) clauses, i.e. a finite verb and its arguments. This is the level of syntactic structure where issues arise of the relative scoping of quantified participants in different roles, as well as relative scoping of event quantification and participant quantification. Annotation structures at this level are quadruples consisting of an event structure, a set of participant structures, a set of link structures that relate participants to events, and a set of link structures that specify scope relations; see (5), where ev is an event structure; P 1 ... P n are participant structures; L P 1 , ..., L P n are participation link structures, and sc 1 , ..., sc k are scope link structures.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "(5) A = ev , { P 1 , ..., P n }, {L P 1 , ..., L P n }, {sc 1 , ..., sc k } 1 This listing is slightly simplified. For the full specification see Bunt (2020).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "QuantML abstract syntax", "sec_num": "2.3" }, { "text": "The design of QuantML was inspired by the theory of generalized quantifiers (GQT, Barwise and Cooper 1981; Keenan and Westerstaahl) , 1997, combined with neo-Davidsonian event semantics (Davidson, 1967; Parsons, 1990) , viewing natural language quantifiers as properties of sets of participants involved in sets of events. Champollion (2015) has shown the viability of this type of combination.", "cite_spans": [ { "start": 82, "end": 106, "text": "Barwise and Cooper 1981;", "ref_id": "BIBREF0" }, { "start": 107, "end": 131, "text": "Keenan and Westerstaahl)", "ref_id": null }, { "start": 186, "end": 202, "text": "(Davidson, 1967;", "ref_id": "BIBREF13" }, { "start": 203, "end": 217, "text": "Parsons, 1990)", "ref_id": "BIBREF22" } ], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "QuantML has an interpretation-by-translation semantics in the form of a compositional, recursive translation of annotation structures to Discourse Representation Structures (DRSs) as defined by (Kamp and Reyle, 1993) . If the annotation structure is fully connected, i.e., if (1) all participant entity structures are linked to an event structure, and (2) for any two participant entity structures linked to the same event structure the relative scopes are specified, then the interpretation function delivers a standard DRS; if one or both of these conditions are not satisfied, then the interpretation delivers an underspecified DRS (UDRS, Reyle, 1984) .", "cite_spans": [ { "start": 194, "end": 216, "text": "(Kamp and Reyle, 1993)", "ref_id": "BIBREF19" }, { "start": 642, "end": 654, "text": "Reyle, 1984)", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "The QuantML semantics is compositional in the sense that the interpretation of an annotation structure is obtained by combining the interpretations of its component entity structures and participation link structures in a manner that is determined by the scope link structures. Combining GQT Casting the semantics in this form is particularly convenient for combining annotations of quantification with other types of semantic information, using annotation schemes of the ISO Semantic Annotation Framework (SemAF) and annotation scheme plug-ins (Bunt, 2019) .", "cite_spans": [ { "start": 545, "end": 557, "text": "(Bunt, 2019)", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "The semantic entities that correspond to participant entity structures may be of any kind: realworld objects, abstract entities, events, individual concepts, intentional and intensional entities, hypothetical and fictional entities. The design of QuantML aims to be neutral with respect to ontological and linguistic views on the existence of objects of various kinds and the need for them in semantic accounts of natural language.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "Note that a participation link structure embeds the linked event structure and participant structure, to the effect that the annotation structures as defined by the abstract syntax are nested structures, as opposed to their flat XML-representations. The interpretation of a fully-connected annotation structure is therefore determined by the interpretation of the participation link structures.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "The semantics of a participation link structure is a combination of the semantics of its components by means of the interpretation function I Q as specified in (6), where \u222a is the operation of merging two DRSs, as defined in DRT, and \u222a * is the scoped merge operation, defined below in (11).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "(6) a.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "I Q ( E , P , R, d, narrow) = (I Q ( P ) \u222a * I Q ( E )) \u222a I Q (R, d, narrow) b. I Q ( E , P , R, d, wide) = (I Q ( E ) \u222a * I Q ( P )) \u222a I Q (R, d, wide) c. I Q ( E , P , R, d, free) = (I Q ( E ) \u222a I Q ( P )) \u222a I Q (R, d, free)", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "As an illustration, consider sentence (7), with its annotation and interpretation shown in Figure 1 . 7All the students read three papers A quantifier of the form \"All the D\" is interpreted as a DRS of the form (8), where capital letters are used for discourse referents that correspond to non-empty sets of individuals. This DRS says that there is a non-empty subset X of the quantification domain D containing all the contextually distinguished students, using the subscript '0' to indicate the contextually determined 'reference domain' or 'context set' (Westerst\u00e5hl, 1985) ). This subset X contains those elements of the reference domain that participate in a set of events. For the quantifier \"All the students\" this leads to the interpretation (8b). Similarly, the annotation of the quantifier \"three papers\" leads to the interpretation (8c).", "cite_spans": [ { "start": 557, "end": 576, "text": "(Westerst\u00e5hl, 1985)", "ref_id": "BIBREF27" } ], "ref_spans": [ { "start": 91, "end": 99, "text": "Figure 1", "ref_id": "FIGREF1" } ], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "(8) a. [X|x \u2208 X \u2194 D 0 (x)] b. [X|x \u2208 X \u2194 student 0 (x)] c. [Y ||Y | = 3, y \u2208 Y \u2192 paper(y) ]", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "For the semantic role R, the distribution d = 'individual', and the event scope \u03c3 = 'narrow', the interpretation of the third component in (6) is the DRS in (9), which says that there is a non-empty participant set of which every member has the role R in a non-empty set of events:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "(9) I Q (R,individual,narrow) = [X|x \u2208 X \u2194 D 0 (x), x \u2208 X \u2192 [E|e \u2208 E \u2192 R(e, x)]]", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "Application of (6) and merging the DRS in (9) with the DRSs interpreting the participant structure and the event structure, results in (10) for the interpretation of the annotation of the sentence in (7).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "(10) [X|x \u2208 X \u2194 student o (x), x \u2208 X \u2192 [Y ||Y | = 3, y \u2208 Y \u2192 [E|paper(y), e \u2208 E \u2192[|agent(e, x), theme(e, y)]]]", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "The scoped merge operation is designed to combine the information about quantified participation in two participation link structures, and is defined as follows:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "(11) The scoped merge operation combines the information in its argument DRSs into a DRS that reflects the relative scoping of the quantifications involved, as well as the relative scopings of participants and events, while unifying the event discourse referents in the two arguments. (If this unification is not possible, then the operation fails. )", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "For annotation structures that do not fully specify the relative scopes of all the sets of participants involved in the same events, the semantic interpretation takes the form of a set of (sub-)DRSs that express the semantics of the participation link structures, plus the scope restrictions for their possible combination. Such an interpretation is known in DRT as an underspecified DRS (UDRS, Reyle, 1994) .", "cite_spans": [ { "start": 395, "end": 407, "text": "Reyle, 1994)", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "A detailed specification of the semantics of QuantML annotation structures can be found in the technical report Bunt (2020) , available on the ISA-17 website https://sigsem.uvt.nl/isa17/ TiCC_Report_Quantification-12-Print.pdf.", "cite_spans": [ { "start": 112, "end": 123, "text": "Bunt (2020)", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "3 The Quantification Challenge test suite", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Semantics", "sec_num": "2.4" }, { "text": "The Quantification Challenge test suite has been constructed in such a way that its sentences illustrate the coverage of the QuantML proposal, with a number of challenging borderline cases that invite speculation and creativeness in finding adequate annotations. More specifically, the test suite covers the following phenomena:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Definiteness and determinacy of NPs. Where an NP like \"the students\" is obviously definite, and semantically determinate, less obvious is how to characterize \"some of the students\" or \"one of my friends\".", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Attributive and predicative adjectives.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Deictic NPs such as \"I\" and \"you\".", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "(7) All the students read three papers.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "Markables: m1 = all the students, m2 = students, m3 = read, m4 = three papers, m5 = papers. m2, student,count , all, det , m4, m5, paper,count , 3, indet }, { m3, read , m1, m2, student, count , all, indet , agent, individual, narrow , { m3, read , m4, m5, paper, count , 3, indet , theme, individual, narrow } m3, read , { m1, m2, student,count , all, det , m4, m5, paper,count , 3, indet , wider c. Semantics: \u2022 Scope ambiguities, as in \"The editors didn't see a misprint\".", "cite_spans": [ { "start": 92, "end": 397, "text": "m2, student,count , all, det , m4, m5, paper,count , 3, indet }, { m3, read , m1, m2, student, count , all, indet , agent, individual, narrow , { m3, read , m4, m5, paper, count , 3, indet , theme, individual, narrow } m3, read , { m1, m2, student,count , all, det , m4, m5, paper,count , 3, indet , wider", "ref_id": null } ], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "A = ev , { P 1 , P 2 }, {L P 1 , L P 2 }, {sc 1 } = = m3, read , { m1,", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "I Q (A) = I Q (L P 1 ) \u222a * I Q (L P 2 ) \u222a * I Q )( ev ) = [X|x \u2208 X \u2194 student 0 (x), x \u2208 X \u2192 [Y |y \u2208 Y \u2192 [E| paper(y), e \u2208 E \u2192 [ agent(e, x", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Conjoined NPs, like \"Bert and Alice\".", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Relative clauses.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Proper names.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Temporal quantifiers, such as \"twice\", \"two to three times\", and \"every hour\".", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Negations.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Mass NPs.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Anaphoric possessive pronouns (\"his\", \"their\").", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Complex possessives, as in \"The headmaster's childrens' toys\".", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Collective quantifications, as in \"These machines combine 12 parts\", interpreted as saying that each of the machines every time combines twelve parts.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Exhasutive quantification.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Quantification with unscpecific distribution, as in \"The boys carried the boxes upstairs\".", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "\u2022 Complex NN-modifications, like \"new corona virus infections\".", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Quantification phenomena", "sec_num": "3.1" }, { "text": "The sentences in the test suite all come with a suggestion for substrings to be used as markables in the annotation. This is to make the comparison of annotations made by different annotators easier, .", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Markables", "sec_num": "3.2" }, { "text": "Concerning the choice of markables for a given (small) clause, first of all every NP is naturally a markable, describing a set of participants (or possibly a single participant), and the main verb (possibly with modifiers) is another markable, corresponding to the events in which the participants are involved. Other markables are those words that correspond to the predicates of the conceptual inventory in the abstract syntax and those in the concrete syntax, notably as values of the @pred attribute. This concerns all nouns, adjectives, prepositions, and numerical as well as non-numerical terms.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Markables", "sec_num": "3.2" }, { "text": "A direct consequence of this way of distinguish-ing markables, is that they may overlap; for example, the markable for an NP overlaps with the one for its head noun. In such a case, the numbering of markables is determined in the first place by its left boundary, and if they have the same left boundary, than by the linear position of the right boundary. So in the sentence \"Most of the students passed the exam\", Most\" is numbered as markable m1, and \"Most of the students\" as markable m2. Markables may be discontinuous, for example, in \"The boys carried the boxes upstairs\", the words \"carried upstairs\" is a discontinuous markable. Following the ordering convention of Discontinuous Phrase Structure Grammar (Bunt, 1996) , the numbering of discontinuous markables is determined by their leftmost element, and if two such markables have the same leftmost element then by their next element, and so on.", "cite_spans": [ { "start": 713, "end": 725, "text": "(Bunt, 1996)", "ref_id": "BIBREF1" } ], "ref_spans": [], "eq_spans": [], "section": "Markables", "sec_num": "3.2" }, { "text": "The documentation for annotating and interpreting the sentences of the test suite, in particular the technical report (Bunt, 2018) , defines concepts and provides guidelines for dealing with these phenomena. These guidelines have not yet been very well developed, and a secondary purpose of the ISA-17 Quantification Challenge, besides the identification of its strengths and weaknesses for annotating quantification phenomena in a semantically adequate way, is to obtain a good picture of the ways in which these guidelines can be improved and extended. The introduction of decision trees to support annotators in choosing the right values of QuantML attributes may for example be an attractive direction.", "cite_spans": [ { "start": 118, "end": 130, "text": "(Bunt, 2018)", "ref_id": "BIBREF6" } ], "ref_spans": [], "eq_spans": [], "section": "Annotation guidelines", "sec_num": "3.3" } ], "back_matter": [], "bib_entries": { "BIBREF0": { "ref_id": "b0", "title": "Generalized Quantifiers and Natural Language. Linguistics and Philosophy", "authors": [ { "first": "J", "middle": [], "last": "Barwise", "suffix": "" }, { "first": "R", "middle": [], "last": "Cooper", "suffix": "" } ], "year": 1981, "venue": "", "volume": "4", "issue": "", "pages": "159--219", "other_ids": {}, "num": null, "urls": [], "raw_text": "J. Barwise and R. Cooper. 1981. Generalized Quanti- fiers and Natural Language. Linguistics and Philos- ophy, 4:159-219.", "links": null }, "BIBREF1": { "ref_id": "b1", "title": "Formal tools for the description and processing of discontinuous constituents", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" } ], "year": 1996, "venue": "Discontinuous Constituency", "volume": "", "issue": "", "pages": "63--85", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt. 1996. Formal tools for the description and pro- cessing of discontinuous constituents. In Harry Bunt and Arthur van Horck, editors, Discontinuous Con- stituency., pages 63-85. Mouton De Gruyter, Berlin.", "links": null }, "BIBREF2": { "ref_id": "b2", "title": "The DIT++ taxonomy for functional dialogue markup", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" } ], "year": 2009, "venue": "Proceedings of AAMAS 2009", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt. 2009. The DIT++ taxonomy for functional dialogue markup. In Proceedings of AAMAS 2009", "links": null }, "BIBREF3": { "ref_id": "b3", "title": "Workshop 'Towards a Standard Markup Language for Embodied Dialogue Acts", "authors": [], "year": null, "venue": "", "volume": "", "issue": "", "pages": "13--24", "other_ids": {}, "num": null, "urls": [], "raw_text": "Workshop 'Towards a Standard Markup Language for Embodied Dialogue Acts', pages 13-24, Bu- dapest.", "links": null }, "BIBREF4": { "ref_id": "b4", "title": "A methodology for designing semantic annotation languages exploring semantic-syntactic ISO-morphisms", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" } ], "year": 2010, "venue": "Proceedings of the Second International Conference on Global Interoperability for Language Resources (ICGL 2010)", "volume": "", "issue": "", "pages": "29--46", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt. 2010. A methodology for designing semantic annotation languages exploring semantic-syntactic ISO-morphisms. In Proceedings of the Second In- ternational Conference on Global Interoperability for Language Resources (ICGL 2010), pages 29-46, Hong Kong: City University.", "links": null }, "BIBREF5": { "ref_id": "b5", "title": "On the principles of semantic annotation", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" } ], "year": 2015, "venue": "Proceedings 11th Joint ACL-ISO Workshop on Interoperable Semantic Annotation (ISA-11)", "volume": "", "issue": "", "pages": "1--13", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt. 2015. On the principles of semantic anno- tation. In Proceedings 11th Joint ACL-ISO Work- shop on Interoperable Semantic Annotation (ISA- 11), pages 1-13, London.", "links": null }, "BIBREF6": { "ref_id": "b6", "title": "Semantic Annotation of Quantification in Natural Language", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" } ], "year": 2018, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt. 2018. Semantic Annotation of Quantification in Natural Language. TiCC Technical Report 2018- 15. Tilburg Center for Cognition and Communica- tion, Tilburg University.", "links": null }, "BIBREF7": { "ref_id": "b7", "title": "An annotation scheme for quantification", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" } ], "year": 2019, "venue": "Proceedings 13th International Conference on Computational Semantics (IWCS 2019)", "volume": "", "issue": "", "pages": "31--43", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt. 2019a. An annotation scheme for quantifica- tion. In Proceedings 13th International Conference on Computational Semantics (IWCS 2019), pages 31-43, University of Gothenburg.", "links": null }, "BIBREF8": { "ref_id": "b8", "title": "Plug-ins for content annotation of dialogue acts", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" } ], "year": 2019, "venue": "Proceedings 15th Joint ISO-ACL Workshop on Interoperable Semantic Annotation (ISA-15)", "volume": "", "issue": "", "pages": "34--45", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt. 2019b. Plug-ins for content annotation of dialogue acts. In Proceedings 15th Joint ISO- ACL Workshop on Interoperable Semantic Annota- tion (ISA-15), pages 34-45, Gothenburg, Sweden.", "links": null }, "BIBREF9": { "ref_id": "b9", "title": "Semantic Annotation of Quantification in Natural Language", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" } ], "year": 2021, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt. 2021. Semantic Annotation of Quantification in Natural Language, second edition. TiCC Techni- cal Report 2021-2. Tilburg Center for Cognition and Communication, Tilburg University.", "links": null }, "BIBREF10": { "ref_id": "b10", "title": "Towards an ISO Standar for the Annotation of Quantification", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" }, { "first": "J", "middle": [], "last": "Pustejovsky", "suffix": "" }, { "first": "K", "middle": [], "last": "Lee", "suffix": "" } ], "year": 2018, "venue": "Proceedings of the 11th International Conference on Language Resources and Evaluation (LREC 2018)", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt, J. Pustejovsky, and K. Lee. 2018. Towards an ISO Standar for the Annotation of Quantification. In Proceedings of the 11th International Conference on Language Resources and Evaluation (LREC 2018), Miyazaki, Japan.", "links": null }, "BIBREF11": { "ref_id": "b11", "title": "Towards Multimodal Content Representation", "authors": [ { "first": "H", "middle": [], "last": "Bunt", "suffix": "" }, { "first": "L", "middle": [], "last": "Romary", "suffix": "" } ], "year": 2002, "venue": "Workshop on International Standards of Terminology and Language Resources Management", "volume": "", "issue": "", "pages": "54--60", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Bunt and L. Romary. 2002. Towards Multimodal Content Representation. In Proceedings of LREC 2002, Workshop on International Standards of Ter- minology and Language Resources Management, pages 54-60, Las Palmas. Paris: ELRA.", "links": null }, "BIBREF12": { "ref_id": "b12", "title": "The interaction of compositional semantics and event semantics", "authors": [ { "first": "L", "middle": [], "last": "Champollion", "suffix": "" } ], "year": 2015, "venue": "Linguistics and Philosophy", "volume": "38", "issue": "1", "pages": "31--66", "other_ids": {}, "num": null, "urls": [], "raw_text": "L. Champollion. 2015. The interaction of composi- tional semantics and event semantics. Linguistics and Philosophy, 38 (1):31-66.", "links": null }, "BIBREF13": { "ref_id": "b13", "title": "The logical form of action sentences", "authors": [ { "first": "D", "middle": [], "last": "Davidson", "suffix": "" } ], "year": 1967, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "D. Davidson. 1967. The logical form of action sen- tences. In N. Rescher, editor, The Logic of Decision and Action. University of Pittsburgh Press.", "links": null }, "BIBREF14": { "ref_id": "b14", "title": "International standard for a linguistic annotation framework", "authors": [ { "first": "N", "middle": [], "last": "Ide", "suffix": "" }, { "first": "L", "middle": [], "last": "Romary", "suffix": "" } ], "year": 2004, "venue": "Natural Language Engineering", "volume": "10", "issue": "", "pages": "221--225", "other_ids": {}, "num": null, "urls": [], "raw_text": "N. Ide and L. Romary. 2004. International standard for a linguistic annotation framework. Natural Lan- guage Engineering, 10:221-225.", "links": null }, "BIBREF15": { "ref_id": "b15", "title": "ISO 24612: Language resource management: Linguistic annotation framework (LAF). International Organisation for Standardisation ISO", "authors": [ { "first": "", "middle": [], "last": "Iso", "suffix": "" } ], "year": 2010, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "ISO. 2010. ISO 24612: Language resource man- agement: Linguistic annotation framework (LAF). International Organisation for Standardisation ISO, Geneva.", "links": null }, "BIBREF16": { "ref_id": "b16", "title": "ISO 24617-1: Language resource management -Semantic annotation framework -Part 1: Time and events ('ISO-TimeML'). International Organisation for Standardisation ISO", "authors": [ { "first": "", "middle": [], "last": "Iso", "suffix": "" } ], "year": 2012, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "ISO. 2012. ISO 24617-1: Language resource man- agement -Semantic annotation framework -Part 1: Time and events ('ISO-TimeML'). International Organisation for Standardisation ISO, Geneva.", "links": null }, "BIBREF17": { "ref_id": "b17", "title": "ISO 24617-2: Language resource management -Semantic annotation framework -Part 4: Semantic roles. International Organisation for Standardisation ISO", "authors": [ { "first": "", "middle": [], "last": "Iso", "suffix": "" } ], "year": 2014, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "ISO. 2014. ISO 24617-2: Language resource man- agement -Semantic annotation framework -Part 4: Semantic roles. International Organisation for Stan- dardisation ISO, Geneva.", "links": null }, "BIBREF18": { "ref_id": "b18", "title": "Language resource management -Semantic annotation framework -Part 6:Principles of semantic annotation. International Organisation for Standardisation ISO, Geneva. ISO. 2016b. ISO WD 24617-2: Language resource management -Semantic annotation framework -Part", "authors": [], "year": null, "venue": "Quantification ('QuantML', Working Draft). International Organisation for Standardisation ISO", "volume": "12", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "ISO. 2016a. ISO 24617-2: Language resource man- agement -Semantic annotation framework -Part 6:Principles of semantic annotation. International Organisation for Standardisation ISO, Geneva. ISO. 2016b. ISO WD 24617-2: Language re- source management -Semantic annotation frame- work -Part 12: Quantification ('QuantML', Work- ing Draft). International Organisation for Standardi- sation ISO, Geneva.", "links": null }, "BIBREF19": { "ref_id": "b19", "title": "From Discourse to Logic", "authors": [ { "first": "H", "middle": [], "last": "Kamp", "suffix": "" }, { "first": "U", "middle": [], "last": "Reyle", "suffix": "" } ], "year": 1993, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "H. Kamp and U. Reyle. 1993. From Discourse to Logic. Kluwer Academic Publishers, Dordrecht.", "links": null }, "BIBREF20": { "ref_id": "b20", "title": "Generalized Quantifiers in Linguistics and Logic", "authors": [ { "first": "E", "middle": [], "last": "Keenan", "suffix": "" }, { "first": "D", "middle": [], "last": "Westerst\u00e5hl", "suffix": "" } ], "year": 1997, "venue": "Generalized Quantifiers in Natural Language", "volume": "", "issue": "", "pages": "837--993", "other_ids": {}, "num": null, "urls": [], "raw_text": "E. Keenan and D. Westerst\u00e5hl. 1997. Generalized Quantifiers in Linguistics and Logic. In General- ized Quantifiers in Natural Language, pages 837- 993. Foris, Dordrecht.", "links": null }, "BIBREF21": { "ref_id": "b21", "title": "SpatialML: Annotation Scheme, Resources, and Evaluation. Language Resources and Evaluation", "authors": [ { "first": "I", "middle": [], "last": "Mani", "suffix": "" }, { "first": "C", "middle": [], "last": "Doran", "suffix": "" }, { "first": "D", "middle": [], "last": "Harris", "suffix": "" }, { "first": "J", "middle": [], "last": "Hitzeman", "suffix": "" }, { "first": "S", "middle": [ "R" ], "last": "Quinby", "suffix": "" }, { "first": "J", "middle": [], "last": "Richer", "suffix": "" }, { "first": "B", "middle": [], "last": "Wellner", "suffix": "" }, { "first": "S", "middle": [], "last": "Mardis", "suffix": "" }, { "first": "S", "middle": [], "last": "Clancy", "suffix": "" } ], "year": 2010, "venue": "", "volume": "44", "issue": "", "pages": "263--280", "other_ids": {}, "num": null, "urls": [], "raw_text": "I. Mani, C. Doran, D. Harris, J.Hitzeman, S.R. Quinby, J. Richer, B. Wellner, S. Mardis, and S. Clancy. 2010. SpatialML: Annotation Scheme, Resources, and Evaluation. Language Resources and Evalua- tion, 44(3):263-280.", "links": null }, "BIBREF22": { "ref_id": "b22", "title": "Events in the Semantics of English: A Study in Subatomic Semantics", "authors": [ { "first": "T", "middle": [], "last": "Parsons", "suffix": "" } ], "year": 1990, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "T. Parsons. 1990. Events in the Semantics of English: A Study in Subatomic Semantics. MIT Press, Cam- bridge, MA.", "links": null }, "BIBREF23": { "ref_id": "b23", "title": "The Semantics of Possessives. Language", "authors": [ { "first": "S", "middle": [], "last": "Peters", "suffix": "" }, { "first": "D", "middle": [], "last": "Westerst\u00e5hl", "suffix": "" } ], "year": 2013, "venue": "", "volume": "89", "issue": "", "pages": "713--759", "other_ids": {}, "num": null, "urls": [], "raw_text": "S. Peters and D. Westerst\u00e5hl. 2013. The Semantics of Possessives. Language, 89(4):713-759.", "links": null }, "BIBREF24": { "ref_id": "b24", "title": "Designing annotation schemes: From theory to model", "authors": [ { "first": "J", "middle": [], "last": "Pustejovsky", "suffix": "" }, { "first": "H", "middle": [], "last": "Bunt", "suffix": "" }, { "first": "A", "middle": [], "last": "Zaenen", "suffix": "" } ], "year": 2017, "venue": "Handbook of Linguistic Annotation", "volume": "", "issue": "", "pages": "21--72", "other_ids": {}, "num": null, "urls": [], "raw_text": "J. Pustejovsky, H. Bunt, and A. Zaenen. 2017. De- signing annotation schemes: From theory to model. In Nancy Ide and James Pustejovsky, editors, Handbook of Linguistic Annotation, pages 21-72. Springer, Berlin.", "links": null }, "BIBREF25": { "ref_id": "b25", "title": "Temporal and event information in natural language text", "authors": [ { "first": "J", "middle": [], "last": "Pustejovsky", "suffix": "" }, { "first": "R", "middle": [], "last": "Knippen", "suffix": "" }, { "first": "J", "middle": [], "last": "Litman", "suffix": "" }, { "first": "R", "middle": [], "last": "Sauri", "suffix": "" } ], "year": 2007, "venue": "Computing Meaning", "volume": "3", "issue": "", "pages": "301--346", "other_ids": {}, "num": null, "urls": [], "raw_text": "J. Pustejovsky, R. Knippen, J. Litman, and R. Sauri. 2007. Temporal and event information in natural language text. In Computing Meaning, Vol. 3, pages 301-346. Springer, Dordrecht.", "links": null }, "BIBREF26": { "ref_id": "b26", "title": "Dealing with ambiguities by underspecification: Construction, representation, and deduction", "authors": [ { "first": "U", "middle": [], "last": "Reyle", "suffix": "" } ], "year": 1993, "venue": "Journal of Semantics", "volume": "", "issue": "", "pages": "123--179", "other_ids": {}, "num": null, "urls": [], "raw_text": "U. Reyle. 1993. Dealing with ambiguities by under- specification: Construction, representation, and de- duction. Journal of Semantics, pages 123-179.", "links": null }, "BIBREF27": { "ref_id": "b27", "title": "Determiners and context sets", "authors": [ { "first": "D", "middle": [], "last": "Westerst\u00e5hl", "suffix": "" } ], "year": 1985, "venue": "Generalized Quantifiers in Natural Language", "volume": "", "issue": "", "pages": "45--71", "other_ids": {}, "num": null, "urls": [], "raw_text": "D. Westerst\u00e5hl. 1985. Determiners and context sets. In Johan van Benthem and Alice ter Meulen, editors, Generalized Quantifiers in Natural Language, pages 45-71. Foris, Dordrecht.", "links": null } }, "ref_entries": { "FIGREF0": { "num": null, "text": "), theme(e, y)]]]]]", "uris": null, "type_str": "figure" }, "FIGREF1": { "num": null, "text": "Example annotation with abstract syntax and semantics", "uris": null, "type_str": "figure" } } } }