Language resource management — Semantic annotation framework (SemAF) — Part 14: Spatial semantics

This document extends ISO 24617-7:2020, which specifies ways of annotating spatial information in natural language such as English, by establishing a formal semantics for its abstract syntax. The task of the proposed semantics is of two kinds: a) translation of annotation structures to semantic forms; b) model-theoretic interpretation of semantic forms. Semantic forms are represented in a type-theoretic first-order logic. These semantic forms are then interpreted with respect to a model for part of the world to which an annotated language is referentially, or denotationally, anchored. NOTE The basic framework and content of this document is based on Reference [1].

Gestion des ressources linguistiques — Cadre d'annotation sémantique (SemAF) — Partie 14: Sémantique spatiale

Le présent document étend l’ISO 24617-7:2020, qui spécifie des manières d’annoter l’information spatiale en langue naturelle, comme l’anglais, en établissant une sémantique formelle pour sa syntaxe abstraite. La tâche de la sémantique proposée est de deux natures: a) transposition des structures d’annotation en formes sémantiques; b) interprétation des formes logiques en théorie des modèles. Les formes sémantiques sont représentées dans une logique du premier ordre de théorie des types. Ces formes sémantiques sont ensuite interprétées par rapport à un modèle d’une partie du monde auquel une langue annotée est rattachée par référence ou par dénotation. NOTE Le cadre de base et le contenu du présent document sont fondés sur la Référence[1].

Upravljanje jezikovnih virov - Ogrodje za semantično označevanje (SemAF) - 14. del: Prostorska semantika

General Information

Status
Published
Publication Date
25-Jun-2023
Current Stage
6060 - International Standard published
Start Date
26-Jun-2023
Due Date
08-Oct-2023
Completion Date
26-Jun-2023

Buy Standard

Standard
ISO 24617-14:2023 - Language resource management — Semantic annotation framework (SemAF) — Part 14: Spatial semantics Released:26. 06. 2023
English language
14 pages
sale 15% off
Preview
sale 15% off
Preview
Draft
ISO/DIS 24617-14:2022 - BARVE
English language
19 pages
sale 10% off
Preview
sale 10% off
Preview
e-Library read for
1 day

Standards Content (Sample)

INTERNATIONAL ISO
STANDARD 24617-14
First edition
2023-06
Language resource management —
Semantic annotation framework
(SemAF) —
Part 14:
Spatial semantics
Gestion des ressources linguistiques — Cadre d'annotation
sémantique (SemAF) —
Partie 14: Sémantique spatiale
Reference number
ISO 24617-14:2023(E)
© ISO 2023

---------------------- Page: 1 ----------------------
ISO 24617-14:2023(E)
COPYRIGHT PROTECTED DOCUMENT
© ISO 2023
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii
  © ISO 2023 – All rights reserved

---------------------- Page: 2 ----------------------
ISO 24617-14:2023(E)
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Metamodel . 2
5 Semantic types . 4
5.1 General . 4
5.2 Basic types . 4
5.2.1 General . 4
5.2.2 Extended basic types . 4
5.2.3 Functional types . 4
5.3 Place and spatial entity . 5
5.4 Paths . 6
6 Events and paths generated from events . 7
6.1 General . 7
6.2 Two types of verb constructions . 7
6.3 Typing event-paths . 8
7 Semantic interpretation of annotation structures . 9
7.1 Overview . 9
7.2 Semantic forms . . . 9
7.3 Model theory . 11
7.3.1 General . 11
7.3.2 Interpretation .12
Bibliography .14
iii
© ISO 2023 – All rights reserved

---------------------- Page: 3 ----------------------
ISO 24617-14:2023(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards
bodies (ISO member bodies). The work of preparing International Standards is normally carried out
through ISO technical committees. Each member body interested in a subject for which a technical
committee has been established has the right to be represented on that committee. International
organizations, governmental and non-governmental, in liaison with ISO, also take part in the work.
ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of
electrotechnical standardization.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the
different types of ISO document should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www.iso.org/directives).
ISO draws attention to the possibility that the implementation of this document may involve the use
of (a) patent(s). ISO takes no position concerning the evidence, validity or applicability of any claimed
patent rights in respect thereof. As of the date of publication of this document, ISO had not received
notice of (a) patent(s) which may be required to implement this document. However, implementers are
cautioned that this may not represent the latest information, which may be obtained from the patent
database available at www.iso.org/patents. ISO shall not be held responsible for identifying any or all
such patent rights.
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO’s adherence to
the World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT), see
www.iso.org/iso/foreword.html.
This document was prepared by Technical Committee ISO/TC 37, Language and terminology,
Subcommittee SC 4, Language resource management.
A list of all parts in the ISO 24617 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www.iso.org/members.html.
iv
  © ISO 2023 – All rights reserved

---------------------- Page: 4 ----------------------
ISO 24617-14:2023(E)
Introduction
This document establishes a semantic ground for supporting ISO 24617-7 (spatial information), which
specifies an abstract syntax for the annotation of spatial information in language. It also specifies a
way of translating the annotation structures generated by the abstract syntax of ISO 24617-7 into well-
formed semantic forms. These semantic forms are represented in a type-theoretic first-order logic and
made interpretable according to a model.
This document:
— validates the abstract specification of ISO 24617-7 for the annotation of spatial information in
language on semantic grounds;
— specifies an interoperable format for interpreting spatial information, both static and dynamic.
Dynamic spatial information involves spatio-temporal information as well as information about motions
in space and time. This document aims at satisfying such needs. An understanding of information
in natural language is necessary for many computational linguistics and artificial intelligence (AI)
applications. An explicit semantics is necessary for the specification provided by ISO 24617-7, as the
representations created in accord with that language will not have a significant impact on AI and
automatic inference without explicit interpretation.
v
© ISO 2023 – All rights reserved

---------------------- Page: 5 ----------------------
INTERNATIONAL STANDARD ISO 24617-14:2023(E)
Language resource management — Semantic annotation
framework (SemAF) —
Part 14:
Spatial semantics
1 Scope
This document extends ISO 24617-7:2020, which specifies ways of annotating spatial information in
natural language such as English, by establishing a formal semantics for its abstract syntax. The task of
the proposed semantics is of two kinds:
a) translation of annotation structures to semantic forms;
b) model-theoretic interpretation of semantic forms.
Semantic forms are represented in a type-theoretic first-order logic. These semantic forms are then
interpreted with respect to a model for part of the world to which an annotated language is referentially,
or denotationally, anchored.
NOTE The basic framework and content of this document is based on Reference [1].
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content
constitutes requirements of this document. For dated references, only the edition cited applies. For
undated references, the latest edition of the referenced document (including any amendments) applies.
ISO 24617-7:2020, Language resource management — Semantic annotation framework — Part 7: Spatial
information
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org/
3.1
annotation structure
information structure created by marking up some linguistic expressions with relevant (semantic)
information
Note 1 to entry: ISO 24617-7:2020, for instance, creates such annotation structures by marking up place names or
motions and their spatial relations with relevant spatial information.
1
© ISO 2023 – All rights reserved

---------------------- Page: 6 ----------------------
ISO 24617-14:2023(E)
3.2
eigenplace
eigenspace
region or path occupied by an object
Note 1 to entry: A region may be considered as a particular finite path matching to an interval [x,x] such that its
start and endpoint match or are identical. In that case, a region is considered as a point.
3.3
event-path
region of space occupied by a mover (moving object) throughout an event
3.4
first-order logic
formal language, artificially built for reasoning, with the values of its terms, particularly variables,
ranging over individual objects only
Note 1 to entry: Second-order variables such as P, which ranges over properties of an individual, are temporarily
introduced to allow the λ-operation in the process of deriving semantic forms (3.7), see 7.2, Note and Example 2,
b) and c).
3.5
interpretation
function that maps a semantic form (3.7) to its denotation
Note 1 to entry: The interpretation function is represented by ⟦ ⟧ and, for each semantic form a, its denotation or
the value of the interpretation, is represented by ⟦σ(a)⟧.
Note 2 to entry: In a model-theoretic semantics, the interpretation function ⟦ ⟧ is constrained by a model M and,
M
for each semantic form a and a model M, such an interpretation is represented by ⟦σ(a)⟧ .
3.6
model M
set-theoretical construct that represents part of the real or possible world denoted by semantic forms
(3.7)
3.7
semantic form
logical form
representation of the semantic content of an annotation structure (3.1) of expressions in natural
language
Note 1 to entry: The semantic form of an annotation structure a is represented by σ(a), where σ is a function that
maps an annotation structure a to a semantic form that carries the semantic content of a.
Note 2 to entry: Semantic forms are often called “logical forms” because semantic forms are represented by a
logical language such as first-order logic (3.4).
3.8
type
semantic type
kind or sort of an object denoted by a linguistic expression
4 Metamodel
This document shall be used together with ISO 24617-7:2020.
The metamodel presented in this clause outlines the basic semantic structure for the abstract syntax of
ISO 24617-7 for easy reference, which specifies an annotation scheme for the markup of spatial relations,
2
  © ISO 2023 – All rights reserved

---------------------- Page: 7 ----------------------
ISO 24617-14:2023(E)
both static and dynamic, as expressed in text and other media. This specification distinguishes the
following six major categories of spatially relevant elements for markup in natural language:
a) spatial entities: natural or artificial locations in the world that include places, paths and event-
paths, as well as individual entities participating in spatial relations;
b) spatial relators (signals): linguistic markers that establish relations between places and spatial
entities;
c) spatial measures: quantitative information associated with spatial entities;
d) events and motions: eventualities either static or dynamic;
NOTE Unlike static eventualities such as referring to states, dynamic eventualities (motions) involve
movement from one location to another triggering a trajectory (event-path).
e) static spatial relations: specific qualitative configurational, orientational and metric relations
between objects;
f) dynamic spatial relations: movement of an object triggered by a motion from one location to
another creating an event-path.
The corresponding metamodel for these categories is represented in Figure 1.
NOTE Source: Reference [2] with some modifications.
Figure 1 — Metamodel
Qualitative spatial link (qslink) and orientation link (olink) each relate one spatial object to another. In
contrast, qslink_e and olink_e relate an eventuality of a special type such as “live” to a location such as
“Boston” with a spatial signal “in”.
These categories are constrained by semantic types. Each of the categories listed in the abstract syntax
of isoSpace is shown to match one of the semantic types defined in Clause 5.
3
© ISO 2023 – All rights reserved

---------------------- Page: 8 ----------------------
ISO 24617-14:2023(E)
5 Semantic types
5.1 General
The semantics of isoSpace is formulated on the basis of its abstract syntax, but its interpretation rules
apply to the semantic forms which are derived from annotation structures as represented by a concrete
syntax. Hence, there are two levels of interpretation that shall be identified when defining a formal
semantics of an annotation structure, as applied to linguistic expressions in natural language:
— language to abstract model;
— concrete model to abstract model.
This clause focuses on the first mapping. It articulates the underlying semantics of the entities
represented in the metamodel in type-theoretic terms and demonstrates the composition of examples
within each category. Clause 6 illustrates the second mapping, from the annotation structure
(implemented as a concrete syntactic expression) into the abstract model.
5.2 Basic types
5.2.1 General
The model-theoretic semantics of ISO 24617-7:2020 is based on a theory of semantic types, which sorts
out various objects denoted by linguistic expressions or their annotation structures. It is assumed that
a model is characterized with the basic types in 5.2.2 and the functional types in 5.2.3, corresponding
generally to the categories in Figure 1. Following Reference [3], the list of basic types is extended to
eight basic types from the two basic types (e, the type of objects, and t, the type of truth values) in
Montague Semantics (see Reference [4]) as given in 5.2.2.
5.2.2 Extended basic types
The basic types are as follows:
a) t, the type of truth values;
b) e, the type of objects (entities);
c) i, the type of time points;
d) p, the type of spatial points;
e) v, the type of events;
f) m, the type of measures;
g) int, the type of intervals;
h) vec, the type of vectors.
Further, following Reference [3], the group operator • (bullet) is introduced, which applies to a type to
form a group type, e.g. the group of points, p•.
5.2.3 Functional types
Additional types can be constructed with conventional binary type constructors: → and ⨉. From these,
the standard set of functional types is defined, e.g. e → t, v → t, p → t. Further, a semi-lattice of types
is defined, where ⊑ is a quasi-ordering on the set of types, such that, for types a, b, c: a ⊑ b and b ⊑
c implies a ⊑ c; and a ⊑ a. This introduces the subtyping relation between types: if a ⊑ b, then a is a
subtype of b.
4
  © ISO 2023 – All rights reserved

---------------------- Page: 9 ----------------------
ISO 24617-14:2023(E)
The following typical functional types are derived with the binary type constructor →:
a) e → t, the type of properties of an individual;
b) p → t, the type r of regions;
c) v → t, the type ε of eventuality descriptors;
d) int → p, the type π of (static) paths;
e) int → vec, the type π of vector-based paths.
v
Following the neo-Davidsonian semantics, “John walks” can be represented as [walk(e) ˄ agent(e,j)]
such that “John” is annotated as being the agent of the event “walk”. Here, the variable e refers to an
eventuality of type v, while the verb is an eventuality descriptor denoting a predicate of type v → t or ε.
The individual constant j referring to John is of type e. As for the type of static paths and vector-based
paths, see 5.4.
5.3 Place and spatial entity
The “PLACE” tag is used for annotating geolocations, such as Germany and Boston, as well as geographic
entities such as lakes and mountains. Further, administrative entities that are registered as geolocations
are also tagged as PLACE, e.g. towns and counties. Hence, in Example 1, the qualitative spatial relation
between the two entities is a relation between places. Both “Gothenburg” and “Sweden” are marked
as PLACE, which is typed as “region”. A region, r, is defined as a set of points, p → t. This differs from
Reference [3], where regions are defined as a subtype of p•, while • is a group operator over basic types,
but either analysis can be adopted for these purposes.
NOTE 1 To differentiate tag names from common nouns, tag names are represented in upper case, e.g. PLACE
is a tag for places like Gothenburg or Sweden.
The spatial words such as “location”, “place”, and “region” are all used equivalently. In annotation, they
are all tagged PLACE. Semantically, they do not denote spatial points of type p. For example, Gothenburg
refers to a location, place or region of type p → t.
Further, a qualitative spatial mereo-topological relation within RCC8 (the Region Connection Calculus 8
qualitative spatial relations, see Reference [5]) is typed as a relation between regions: i.e. qslink: r → (r
→ t) for qualitative spatial link, formulated in ISO 24617-7:2020.
EXAMPLE 1
a) [Gothenburg] is [in] [Sweden] .
pl1 s1 pl2
b) ⟦Gothenburg⟧ = G, < G: p→ t>
c) ⟦Sweden⟧ = S, < S: p→ t >
d) ⟦in⟧ = λyλx[in(x,y)], < in: r → (r → t)>
e) in(G, S)
For many spatial relations in language, however, the entities involved are not inherently typed as
locations or places. For example, humans and everyday objects carry a primary type of e, which are
subtyped or identified in these documents as spatialEntity. When they participate in spatial relations,
a type coercion function, 𝓛, is assumed to operate over an entity (or a collection of entities) and returns
the spatial region associated with that entity (or entities), i.e. its location in space. The type for this
localization operator, 𝓛, is: e → (p → t).
5
© ISO 2023 – All rights reserved

---------------------- Page: 10 ----------------------
ISO 24617-14:2023(E)
Example 2 demonstrates how this operator coerces an entity to the type required by the spatial
relation, namely r.
EXAMPLE 2
a) [Robin] in [Sweden] .
se1 p
b) ⟦Robin⟧ = R, < R: e >
c) ⟦Sweden⟧ = S, < S: p→ t >
d) 𝓛(R) = λx[loc(x,R)], < x: p, 𝓛: e → (p →
...

SLOVENSKI STANDARD
oSIST ISO/DIS 24617-14:2022
01-december-2022
Upravljanje jezikovnih virov - Ogrodje za semantično označevanje (SemAF) - 14.
del: Prostorska semantika
Language resource management — Semantic annotation framework (SemAF) — Part
14: Spatial semantics
Gestion des ressources linguistiques — Cadre d'annotation sémantique (SemAF) —
Partie 14: Sémantique spatiale
Ta slovenski standard je istoveten z: ISO/DIS 24617-14:2022
ICS:
01.020 Terminologija (načela in Terminology (principles and
koordinacija) coordination)
01.140.20 Informacijske vede Information sciences
35.240.30 Uporabniške rešitve IT v IT applications in information,
informatiki, dokumentiranju in documentation and
založništvu publishing
oSIST ISO/DIS 24617-14:2022 en,fr,de
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

---------------------- Page: 1 ----------------------
oSIST ISO/DIS 24617-14:2022

---------------------- Page: 2 ----------------------
oSIST ISO/DIS 24617-14:2022
DRAFT INTERNATIONAL STANDARD
ISO/DIS 24617-14
ISO/TC 37/SC 4 Secretariat: KATS
Voting begins on: Voting terminates on:
2022-10-20 2023-01-12
Language resource management — Semantic annotation
framework (SemAF) —
Part 14:
Spatial semantics
Gestion des ressources linguistiques — Cadre d'annotation sémantique —
Partie 14: Sémantique spatiale
ICS: 01.020
THIS DOCUMENT IS A DRAFT CIRCULATED
FOR COMMENT AND APPROVAL. IT IS
THEREFORE SUBJECT TO CHANGE AND MAY
This document is circulated as received from the committee secretariat.
NOT BE REFERRED TO AS AN INTERNATIONAL
STANDARD UNTIL PUBLISHED AS SUCH.
IN ADDITION TO THEIR EVALUATION AS
BEING ACCEPTABLE FOR INDUSTRIAL,
TECHNOLOGICAL, COMMERCIAL AND
USER PURPOSES, DRAFT INTERNATIONAL
STANDARDS MAY ON OCCASION HAVE TO
BE CONSIDERED IN THE LIGHT OF THEIR
POTENTIAL TO BECOME STANDARDS TO
WHICH REFERENCE MAY BE MADE IN
Reference number
NATIONAL REGULATIONS.
ISO/DIS 24617-14:2022(E)
RECIPIENTS OF THIS DRAFT ARE INVITED
TO SUBMIT, WITH THEIR COMMENTS,
NOTIFICATION OF ANY RELEVANT PATENT
RIGHTS OF WHICH THEY ARE AWARE AND TO
PROVIDE SUPPORTING DOCUMENTATION. © ISO 2022

---------------------- Page: 3 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
DRAFT INTERNATIONAL STANDARD
ISO/DIS 24617-14
ISO/TC 37/SC 4 Secretariat: KATS
Voting begins on: Voting terminates on:

Language resource management — Semantic annotation
framework (SemAF) —
Part 14:
Spatial semantics
Gestion des ressources linguistiques — Cadre d'annotation sémantique —
Partie 14: Sémantique spatiale
ICS: 01.020
COPYRIGHT PROTECTED DOCUMENT
THIS DOCUMENT IS A DRAFT CIRCULATED
FOR COMMENT AND APPROVAL. IT IS
© ISO 2022
THEREFORE SUBJECT TO CHANGE AND MAY
This document is circulated as received from the committee secretariat.
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
NOT BE REFERRED TO AS AN INTERNATIONAL
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on STANDARD UNTIL PUBLISHED AS SUCH.
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
IN ADDITION TO THEIR EVALUATION AS
or ISO’s member body in the country of the requester. BEING ACCEPTABLE FOR INDUSTRIAL,
TECHNOLOGICAL, COMMERCIAL AND
ISO copyright office
USER PURPOSES, DRAFT INTERNATIONAL
CP 401 • Ch. de Blandonnet 8
STANDARDS MAY ON OCCASION HAVE TO
BE CONSIDERED IN THE LIGHT OF THEIR
CH-1214 Vernier, Geneva
POTENTIAL TO BECOME STANDARDS TO
Phone: +41 22 749 01 11
WHICH REFERENCE MAY BE MADE IN
Reference number
Email: copyright@iso.org
NATIONAL REGULATIONS.
Website: www.iso.org ISO/DIS 24617-14:2022(E)
RECIPIENTS OF THIS DRAFT ARE INVITED
Published in Switzerland
TO SUBMIT, WITH THEIR COMMENTS,
NOTIFICATION OF ANY RELEVANT PATENT
RIGHTS OF WHICH THEY ARE AWARE AND TO
ii
  © ISO 2022 – All rights reserved
PROVIDE SUPPORTING DOCUMENTATION. © ISO 2022

---------------------- Page: 4 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Overview . 2
4.1 Purpose and justification . 2
4.2 Metamodel . 2
5 Semantic types . 4
5.1 General . 4
5.2 Basic types . 4
5.2.1 General . 4
5.2.2 Extended Basic types . 4
5.2.3 Functional types . 4
5.3 Place and spatial entity . 5
5.4 Paths . 6
6 Events and paths generated from events . 7
6.1 General . 7
6.2 Two types of verb constructions . 7
6.3 Typing event-paths . 8
7 Semantic interpretation of annotation structures . 8
7.1 Two roles . 8
7.2 Semantic forms . . . 9
7.3 Model Theory . 11
7.3.1 General . 11
7.3.2 Basic types . 11
7.3.3 Functional types . 11
7.3.4 Interpretation . 11
Bibliography .14
iii
© ISO 2022 – All rights reserved

---------------------- Page: 5 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards
bodies (ISO member bodies). The work of preparing International Standards is normally carried out
through ISO technical committees. Each member body interested in a subject for which a technical
committee has been established has the right to be represented on that committee. International
organizations, governmental and non-governmental, in liaison with ISO, also take part in the work.
ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of
electrotechnical standardization.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2. www.iso.org/directives
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. ISO shall not be held responsible for identifying any or all such patent rights. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received. www.iso.org/patents
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation on the meaning of ISO specific terms and expressions related to conformity
assessment, as well as information about ISO's adherence to the WTO principles in the Technical
Barriers to Trade (TBT) see the following URL: Foreword - Supplementary information
The committee responsible for this document is ISO/TC 37, Language and Terminology, Subcommittee
SC 4, Language resource management
ISO 24617 consists of the following parts under the general title Language resource management —
Semantic annotation framework (SemAF):
— Part 1: Time and events (SemAF-Time, TimeML)
— Part 2: Dialogue acts (DA)
— Part 3: Named entity
— Part 4: Semantic roles (SR)
— Part 5: Discourse structures (DS)
— Part 6: Principles of semantic annotation (SemAF Principles)
— Part 7: Spatial information
— Part 8: Semantic relations in discourse, core annotation schema (DR-core)
— Part 9: Reference annotation framework (RAF)
— Part 10: Visual information (VoxML)
— Part 11: Measurable quantitative information (MQI)
— Part 12: Quantification
— Part 13: Gestures
— Part 14: Spatial semantics
iv
  © ISO 2022 – All rights reserved

---------------------- Page: 6 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
Introduction
This document provides a semantic ground for supporting ISO 24617-7 Spatial information, which
provides an abstract syntax for the annotation of spatial information in language. It also provides a
way of translating the annotation structures generated by the abstract syntax of ISO 24617-7 into well-
formed logical forms. These logical forms are to be represented in a type-theoretic first-order predicate
logic and made interpretable according to a model.
This document needs to be an ISO standard for at least two reasons. First, it validates the abstract
specification of ISO 24617-7 for the annotation of spatial information in language on semantic grounds.
Second, it provides an interoperable format for interpreting spatial information, both static and
dynamic. Dynamic spatial information involves spatio-temporal information as well as information
about motions in space and time. This document aims at satisfying such needs.
v
© ISO 2022 – All rights reserved

---------------------- Page: 7 ----------------------
oSIST ISO/DIS 24617-14:2022

---------------------- Page: 8 ----------------------
oSIST ISO/DIS 24617-14:2022
DRAFT INTERNATIONAL STANDARD ISO/DIS 24617-14:2022(E)
Language resource management — Semantic annotation
framework (SemAF) —
Part 14:
Spatial semantics
1 Scope
This document provides a formal semantics for the abstract syntax of ISO 24617-7 that specifies ways of
annotating spatial information in natural language such as English. The task of the proposed semantics
is of two kinds: (1) translation of annotation structures to logical forms and (2) model-theoretic
interpretation of logical forms. Logical forms are represented in a type-theoretic first-order predicate
logic (FOL). These logical forms are then interpreted with respect to a model for part of the world to
which an annotated language is referentially, or denotationally, anchored.
[1]
NOTE The basic framework and content of this document is based on, a paper presented at the ISO/TC 37/SC
4/WG 2 workshop, 2019, Santa Fe, U.S.A.
2 Normative references
The following documents, in whole or in part, are normatively referenced in this document and are
indispensable for its application. For dated references, only the edition cited applies. For undated
references, the latest edition of the referenced document (including any amendments) applies.
ISO 24617-7:2020, Language resource management — Semantic annotation framework — Part 7: Spatial
information
3 Terms and definitions
3.1
annotation structure a
information structure created by marking up some linguistic expressions with relevant (semantic)
information
Note 1 to entry: ISO 24617-7:2020, for instance, creates such annotation structures by marking up place names or
motions and their spatial relations with relevant spatial information.
3.2
convex hull of a shape
smallest convex set (3.3) that contains it
3.3
convex set
subset of a Euclidean space such that, given any two points, it contains the whole line segment that joins
them
3.4
eigenplace
eigenspace
region or path occupied by an object
Note 1 to entry: A region may be considered as a particular finite path matching to an interval [x,x] such that its
start and endpoint match or are identical.
1
© ISO 2022 – All rights reserved

---------------------- Page: 9 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
3.5
event-path
region of space occupied by a mover (moving object) throughout an event
3.6
first-order logic
first-order predicate logic with quantification
FOL
formal language, artificially built for reasoning, with the values of its terms, particularly variables,
ranging over individuals only
Note 1 to entry: Second-order variables like P, which ranges over properties of an individual, are temporarily
introduced to allow the λ-operation in the process of deriving logical forms (see Example 6).
3.7
M
interpretation σ(a)
valuation of a logical form σ(a) (3.8) with respect to a model M (3.9)
3.8
logical form σ(a)
representation of the semantic content of an annotation structure a (3.1) of expressions in natural
language
3.9
model M
set-theoretical construct that represents part of the real or possible world denoted by the logical forms
(a) (3.8)
4 Overview
4.1 Purpose and justification
An understanding of information in natural language is necessary for many computational linguistics
and artificial intelligence applications. An explicit semantics is necessary for the specification provided
by ISO 24617-7 Spatial information, as the representations created in accord with that language will
not have a significant impact on artificial intelligene (AI) and automatic inferencing without explicit
interpretation.
Relevant affected stakeholder categories with expected benefits and impacts include: Amazon Echo,
Microsoft, Google Cloud Computing and other related industrial and commerce sectors with geo-
location applications, navigation, home appliance reasoning, and Internet of Things, organizations
like MITRE providing semantic interoperability with natural language processing (NLP) applications,
Government organizations like DARPA, ARL, iARPA, ONR, NSF, and National Geospatial Agency with
impacts on interpretation layer for interoperability between standards used in spatial platforms.
Academic and research bodies such as Brandeis University, Stanford University, KAIST or Korea
University will benefit from spatial metadata enhancement for library special collections and archives
and other managements of bigdata. Even on-governmental organizations like WGBH get help with
spatial metadata indexing over their audio-video archival assets.
4.2 Metamodel
This document outlines the basic semantic structure for the abstract syntax of ISO 24617-7 for easy
reference, which specifies an annotation scheme for the markup of spatial relations, both static and
2
  © ISO 2022 – All rights reserved

---------------------- Page: 10 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
dynamic, as expressed in text and other media. This specification distinguishes six major categories of
spatially relevant elements for markup in natural language:
(1) Categories of markable expressions in ISO 24617-7:
a. spatial entities:
natural or artificial locations in the world that include places, paths, and event-paths, as well
as individual entities participating in spatial relations.
b. spatial relators (signals):
linguistic markers that establish relations between places and spatial entities.
c. spatial measures:
quantitative information associated with spatial entities
d. events and motions:
eventualities either static or dynamic. Unlike static eventualities such as referring to states,
dynamic eventualities (motions) involve movement from one location to another triggering a
trajectory (event-path).
e. static spatial relations:
specific qualitative configurational, orientational, and metric relations between objects.
f. dynamic spatial relations:
movement of an object triggered by a motion from one location to another creating an event
path.
The corresponding metamodel for these categories is represented in Figure 1 below.
Figure 1 — Metamodel
[2]
NOTE 1 Figure 1 is taken from with some modifications.
3
© ISO 2022 – All rights reserved

---------------------- Page: 11 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
NOTE 2 Qualitative spatial link (qslink) and orientation link (olink) relates one spatial object to another,
while these links may relate an eventuality of a special type such as “live” to a location such as “Boston” with a
spatial signal “in”. These relations, represented as qslink_e and olink_e, may be differentiated from ordinary
qslink and olink.
These categories are constrained by semantic types. Each of the categories listed in the abstract syntax
of isoSpace is shown to match one of the semantic types defined in Clause 5.
5 Semantic types
5.1 General
The semantics of isoSpace is formulated on the basis of its abstract syntax, but its interpretation rules
apply to the semantic forms which are derived from annotation structures as represented by a concrete
syntax. Hence, there are two levels of interpretation that need to be identified when defining a formal
semantics of an annotation structure, as applied to linguistic expressions in natural language: language
to abstract model; and concrete model to abstract model.
This clause focuses on the first mapping and articulate the underlying semantics of the entities
represented in the metamodel in type-theoretic terms and demonstrate the composition of examples
within each category. Clause 6 will illustrate the second mapping, from the annotation structure
(implemented as a concrete syntactic expression) into the abstract model.
5.2 Basic types
5.2.1 General
The model-theoretic semantics of isoSpace is based on a theory of semantic types. It is assumed that a
model is characterized with the basic types in 5.2.2 and the functional types in 5.2.3, corresponding
[3]
generally to the categories in Figure 1 above. Following, the list of basic types is extended to 8 basic
types from the two basic types e, the type of objects, and t, the type of truth values, in Montague
[4]
Semantics ( ) as clause in 5.2.2.
5.2.2 Extended Basic types
a. t, the type of truth values.
b. e, the type of objects (entities)
c. i, the type of time points
d. p, the type of spatial points
e. v, the type of events
f. m, the type of measures
g. int, the type of intervals [0,1]
h. vec, the type of vectors
[3]
Further, following again, the group operator • (bullet) is introduced, which applies to a type to form a
group type, e.g., the group of points, p•.
5.2.3 Functional types
Additional types can be constructed with conventional binary type constructors, → and ⨉. From these,
the standard set of functional types is defined: e.g., e → t, v → t, p → t, and so on. Further, a semi-lattice
of types is defined, where ⊑ is a quasi-ordering on the set of types, such that, for types a, b, c: a ⊑ b and
4
  © ISO 2022 – All rights reserved

---------------------- Page: 12 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
b ⊑ c implies a ⊑ c; and a ⊑ a. This introduces the subtyping relation between types: if a ⊑ b, then a is a
subtype of b.
Here are some typical functional types, derived with the binary type contructor →:
a. e → t, the type of properties of an individual
b. p → t, the type r of regions
c. v → t, the type ε of eventutuality descriptors
d. int → p, the type π of (static) paths
e. int → vec, the type πv of vector-based paths
Following the neo-Davidsonian semantics, “John walks” can be represented as [walk(e) ˄ agent(e,j)]
such that “John” is annotated as being the agent of the event walk. Here, the variable e refers to an
eventuality of type v, while the verb is an eventuality descriptor denoting a predicate of type v → t or
ε. The individual constant j referring to John is of type e. As for the the type of static paths and vector-
based paths, please refer to Clause 5.4.
5.3 Place and spatial entity
The place tag is used for annotating geolocations, such as “Germany” and “Boston”, as well as
geographic entities such as lakes and mountains. Further, administrative entities that are registered
as geolocations are also tagged as place, e.g., towns and counties. Hence, in EXAMPLE 1, the qualitative
spatial relation between the two entities is a relation between places. Both “Gothenburg” and “Sweden”
are marked as places, which we will type as regions. A region, r, will be defined as a set of points, p → t.
[3]
This differs from, where regions are defined as a subtype of p•, where • is a group operator over basic
types, but either analysis could be adopted for our present purposes.
Further, a qualitative spatial mereo-topological relation within RCC8 (the Region Connection Calculus 8
qualitative spatial relations, see [5]) will be typed as a relation between regions: i.e., qslink: r → (r → t)
for qualitative spatial link, formulated in ISO 24617-7:2020.
EXAMPLE 1
a. [Gothenburg]pl1 is [in]s1 [Sweden]pl2.
b. Gothenburg = G,
c. Sweden = S,
d. in = λyλx[in(x,y)],
e. in(G, S)
For many spatial relations in language, however, the entities involved are not inherently typed as
locations or places. For example, humans and everyday objects carry a primary type of e, which are
subtyped or identified in these documents as spatialEntity. When they participate in spatial relations,
a type coercion function, 𝓛, is assumed to operate over an entity (or a collection of entities) and returns
the spatial region associated with that entity (or entities), i.e., its location in space. The type for this
localization operator, 𝓛, is: e → (p → t).
EXAMPLE 2 demonstrates how this operator coerces an entity to the type required by the spatial relation,
namely r.
EXAMPLE 2
a. [Robin]se1 in [Sweden]p.
b. Robin = R,
5
© ISO 2022 – All rights reserved

---------------------- Page: 13 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
c. Sweden = S,
d. 𝓛(R) = λx[loc(x,R)],
e. in = λyλx[in(x,y)],
f. in(λx[loc(x,R)], S)
The interpretation of spatialEntity in terms of its localization will hold for how objects participate in c
paths as is shown below.
5.4 Paths
A path is defined as a subtype of locations (formally regions) that have the additional constraint of
being directional and are often construed as one-dimensional.
The notion of a path being introduced or created by an event has its origin in several previous authors,
[6] [7] [8] [7]
including, , and. More recently and more in line with the present specification, the analysis of
is adopted for the specification in isoSpace. Formally, paths have been analyzed as sequences of spaces
[6] [8] [6]
( ) and sequences of vectors. Following, let int be the type of the interval [0,1] ⊂ R, and p be the
type of a spatial point, as defined above. Then a path, π, will be that function int → p, which indexes
locations on the path to values from the interval [0,1]. Similarly, if vec is the type of vectors, then a
vector-based path, πv, can be defined as the function int → vec. That is, it indexes the vectors associated
with the path to values from the interval [0,1].
EXAMPLE 3
a. [Prague]pl1 is on [the Moldau River]p1.
b. [Boston]pl2 is at the end of [the Mass. Turnpike]p2.
In these examples in EXAMPLE 3, the qualitative spatial relation introduced by the predication identifies
a place as situated within (or on) a path. Hence, the preposition “on” which governs the path-PP, [pp on
[NP the Moldau River]], carries a more specific type than a general qslink relation, namely: πv → (r →
t). The type derivation for EXAMPLE 3a is illustrated below.
EXAMPLE 4
a. [Prague]pl1 is on [the Moldau River]p1.
b. Prague = P,
c. the Modau River = M,
d. on = λyλx[onPath(x,y)],
e. onPath(P, M)
As EXAMPLE 3b illustrates, the endpoints of paths can be explicitly mentioned in text. The isoSpace
annotated examples below demonstrate reference to both endpoints and mid-points.
EXAMPLE 5
a. … the railroadpl between Bostonpl1 and [New York]pl2 …
Annotation structure: path(p1, beginID=pl1, endID=pl2, form=NOM)
b. John took the roadp2 through Bostonpl1.
Annotation structures: path(p2, midIDs=pl1, form=NOM)
Formally, the expressions introducing end- and mid-point locations are acting as functions from paths
to path positions: πv → int; e.g., given a path <3,4,5,2,1,8>, end(πv) = 8.
EXAMPLE 6
6
  © ISO 2022 – All rights reserved

---------------------- Page: 14 ----------------------
oSIST ISO/DIS 24617-14:2022
ISO/DIS 24617-14:2022(E)
a. [Boston]pl1 is at the end of [the Mass. Turnpike]p1
b. Boston = B,
c. the Mass. Turnpike = MT,
d. end = λx[endOf(x)],
e. on = λyλx[onPath(x,y)],
f. onPath(B, MT) ˄ endOf(MT) = B
As mentioned above, the localized spatialEntity can be situated on a path by coercion: namely, 𝓛
coerces an entity referred to by the name “John” to his localized place or path, called eigenplace, and
then the spatial relation predication situates this region onto the path, πv.
EXAMPLE 7
a. [John]se1 is on [the road]p1.
b. 𝓛(J) = λx[loc(x, J)],
6 Events and paths generated from events
6.1 General
The term event as it is used in ISO 24167-7:2020 is borrowed directly from ISO 24617-1: 2012 and is
[9] [10]
used as a cover term for situations that happen, occur, hold, or take place. Following and, we can
represent the event as an individual predicated of an event clas
...

Questions, Comments and Discussion

Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.