=Paper= {{Paper |id=Vol-2050/creol-paper1 |storemode=property |title=Natural Language Template Selection for Temporal Constraints |pdfUrl=https://ceur-ws.org/Vol-2050/CREOL_paper_1.pdf |volume=Vol-2050 |authors=C. Maria Keet |dblpUrl=https://dblp.org/rec/conf/jowo/Keet17 }} ==Natural Language Template Selection for Temporal Constraints== https://ceur-ws.org/Vol-2050/CREOL_paper_1.pdf

Natural Language Template Selection for
Temporal Constraints
C. Maria KEET 1
Department of Computer Science, University of Cape Town, South Africa

Abstract. Representing temporal knowledge and information in tempo-
ral logics for ontologies and conceptual data models has faced issues due
to inaccessibility of the underlying logic and limited intuitiveness of di-
agrammatic extensions to the modelling languages. We aim to address
this by designing controlled natural language templates for generating
sentences that verbalise in English the temporal constraints defined in
a temporal logic. We devised 101 templates, which were evaluated by
experts in temporal logics and by novice temporal modellers on seman-
tic adequacy and preference. There was only 12% unanimity among the
experts, and 89% by majority voting. The novice temporal modellers
were much more lenient in judgment on whether the templates cap-
tured the semantics adequately. Instead of a direct 1:1 mapping between
an axiom’s components and the natural language rendering, the more
natural-sounding sentences were preferred, therewith linking an axiom
type as a whole to a template.
Keywords. Temporal logics, Temporal ontologies, Controlled Natural
Language, Temporal conceptual models

1. Introduction

Time is pervasive in communication and relevant for almost any subject domain
of interest. For instance, a business rule that states that any manager of a com-
pany must already be an employee of that company, biological knowledge that
each butterfly used to be a caterpillar, and census information stating that a di-
vorce can only occur if there was a marriage before. Several options are at one’s
disposal to represent such temporal information, be this for ontology development
or information system design. For ontology, the focus has been on fundamental
representation choices such as 3-dimensional objects with time vs 4-dimensional
entities. For representation languages, the emphasis is on features to represent
more or less temporal constraints and automated reasoning over it, which can
be grouped into ontology languages [2,3,9,22,24] and popular conceptual data
modelling languages, such as the UML Class Diagram, Entity-Relationship, and
Object-Role Modeling languages [2,4,10,12,14,17,18,21,23]. These works concern
steps before involving modellers and domain experts. However, given that mod-
1 Corresponding Author: C. Maria Keet, Department of Computer Science, University of Cape

Town, Cape Town, South Africa; E-mail: mkeet@cs.uct.ac.za.
ellers have great difficulty with using such temporal representation languages [15],
then these advances are unlikely to be used. The two principal approaches to ad-
dress this problem is to use a graphical notation or natural language sentences. To
illustrate: a sample model in the TREND language [15] is shown in Figure 1-A.
This is not immediately clear to anyone unless trained in the semantics of the tem-
poral adornments (such as “Dex”; explained further below) [26]. Its logic-based
reconstruction in the DLRU S Description Logic (DL) [2], shown in Figure 1-B,
is typically even less accessible to modellers. The remaining option is a natural
language rendering, which is what we focus on in this paper.

A. Sample temporal model in TREND notation B. Partial logic-based reconstruction
in DLR notation
EmpID Office us
(0,n) (1,n)
Employee work Project

- (1,1)
DEX DEX DEX

Board manage
Manager (1,n)
Member
-
DEV
C. What we aim for: natural language rendering of temporal constraints

Figure 1. Example of a temporally extended ER diagram in TREND notation (A), with a subset
of the corresponding DLRU S axioms as logic-based reconstruction (B), and the verbalised
temporal constraints of the model that resulted from this research (C).

The natural language option means to verbalise structured input or use Nat-
ural Language Generation (NLG). Atemporal verbalisation has been successful
for both conceptual data models [6,13] and ontologies [1,5,7,25,27]; for instance,
for the axiom M anager v Employee and a template Each [subclass] is a(n)
[superclass] for simple subsumption, the verbalisation will generate Each Man-
ager is an Employee. It is an obvious step to seek to use such a template-based
approach also for the temporal constraints. Existing works only generate text
from temporal data rather than information and knowledge; e.g., sample data for
ORM diagrams [8], time series data [19], and querying temporal databases [20].
The aim is thus to find out what is the best way of verbalising the temporal
constraints and answer 1) Does each proposed natural language sentence template
capture the semantics of the temporal constraint adequately? and 2) Which sen-
tence among the options is preferred? We used the semantics of the Description
logic (DL) DLRU S as structured information and knowledge representation, as
it currently has the most comprehensive set of temporal constraints and it eas-
ily can be mapped to one’s preferred logic or modelling language. We discuss
the template development to verbalise the 34 temporal constraints. One or more
templates were devised for each temporal constraint. The templates were evalu-
ated by three temporal logic experts and five ‘mixed experts’ (experts in mod-
elling, logic, or NLG, but not temporal). There was little unanimous agreement
on template preferences among the experts. The ‘mixed experts’ judged many
more sentences to be correct than the experts, yet there was even less agreement
among the participants. Overall, 26 of the 34 constraints did result in preferred
sentences, and the remaining 8 sentences were updated by taking into account
the written feedback. These final templates map an axiom type as a whole to the
template as a whole, rather than by consituent.
In the remainder of the paper, we first describe preliminaries in Section 2.
Possible templates with word choices are discussed in Section 3. The evaluation
is presented in Section 4. We discuss and conclude in Sections 5 and 6.

2. Preliminaries: language and time

Linguistically, there are two principal ways to refer to temporal points and in-
tervals in a sentence: using verb tenses, such as ‘have eaten’ and ‘will eat’, or
prepositions and adjuncts, such as ‘before’ and ‘some time later’, or both to stress
the time dimension and relation between events at points or during intervals, like
‘have eaten before [some other event]’. The major languages in the world have
between zero (e.g., Chinese) and three (e.g., Romance languages) tenses, with
variations using compound forms with auxiliary verbs.
Within commonly used information and knowledge representation that we
restrict ourselves to here, not all verb tenses and adjuncts will be needed, because
it is a restricted application domain concerning tracking the evolution of objects,
relations, and, possibly, attributes over time, not, say, events that did not hap-
pen in the past (‘could/would have done’). The core relevant temporal aspects
are depicted in Figure 2. First, we have “snapshot (rigid)”, which is essentially
atemporal, or ‘at all times’ for the duration of the entity’s existence. The “tem-
poral options (antirigid)” denotes that an entity is an instance of a type at some
time, but not at all times, which is indicated with a non-solid line. There are
four core transition constraints: “extension” means that at some time an entity
also instantiates another type and “evolution” means that the entity instantiates
one type after another. These constraints may be quantitative (a fixed amount of
time), and are either optional (universally quantified) or mandatory (existentially

now
flow of time
snapshot (rigid)
} temporal options (antirigid)
extension in the future
evolution in the future
extension in the past
evolution in the past

Figure 2. Flow of (linear) time with the main options included in temporal conceptual data
models (based on [4]). The horizontal lines depict an object’s life as a member of an entity type,
a tuple (relation) as a member of a relationship, or an attribute.
quantified). As they can be declared for entity types, relationships, and attributes,
this amounts to 48 possible temporal constraints. These changes can be covered
by just a few verb tenses. Mainly, they constitute choices on the future single vs
future continuous (respectively, the past).
Some of the temporal constraints are computationally more interesting than
others. In particular, constraints on the future are trivially satisfied and thus
cannot be checked, whereas constraints about the past—which must be present in
a prior state of the data or knowledge base—and quantitative constraints can be
checked. Also, it was difficult to devise good examples for some of the attribute
constraints, and they are not used often in ontologies anyway. Taking this into
account, we reduced it to 34 axiom types out of the 48. Further, it makes sense to
include domain and range when verbalising a relationship of a conceptual model.
This context is necessary because a conceptual model may reuse the name of an
attribute or relationship, but never with the same entity types. This is in contrast
with relations in ontologies that do not require a domain and range restriction,
and as also reflected in DLRU S . We have made templates for both scenarios,
but it appeared that the templates for the relations (OWL object properties, DL
roles) in ontologies always resulted in a subset of those of conceptual models such
that it required one conversion rule (illustrated below). Therefore, we opt here
for the more comprehensive ones with domain and range.

3. Specific template options

We created 1-7 candidate templates for each constraint, which would verbalise
the constraints more or less precisely or colloquially. For each list of options for a
constraint, the first one is most literal with respect to the formal semantics, and
the other ones are more or less precise rewordings that sound less ‘clunky’ and
artificial. Regarding the formal counterpart, we rely on the formal foundations of
TREND, being the semantics of DLRU S [2], so as not to clutter the paper with
too much repetition and syntax notation. The semantics here and in the following
subsections is based on [4] that was extended with temporal relationships [14],
and the additional mandatory constraints are adapted from [4]. Considering the
usual model-theoretic semantics, we use a temporal interpretation of the signature
of a conceptual data model M. This is a structure of the form: I = (Z, <
), ∆I , {·I(t) | t ∈ Z} , where (Z, <) is the set of integers denoting the intended

flow of time, ∆I 6= ∅ is the interpretation domain divided into ∆IC over classes
and ∆ID over data types, and ·I(t) , for t ∈ Z, is the interpretation function which
assigns a set C I(t) ⊆ ∆I to each entity type C ∈ C, a set RI(t) of tuples over
∆IC × ∆IC to each relation R ∈ R and a set AI(t) of tuples over ∆IC × ∆ID to each
attribute A ∈ A. While DLRU S permits n-ary relations, we present just the case
for binaries. The formalisation of all constraints is available in the supplementary
material at http://www.meteck.org/files/CREOL17suppl.zip.

3.1. Model elements

The entity types, relationships, and attributes can be divided into atemporal
(‘snapshot’) and temporal entities, where the former has no explicit specification
of time—or: holds globally at all times—and the latter do. A temporal entity type
0
is formalised as o ∈ C I(t) → ∃t0 6= t.o ∈/ C I(t ) . A template option with a closer
1:1 match between the axiom type with its structure and the natural sentence
and, say, the entity type Student would generate If an object is an instance of en-
tity type Student, then there is some time where it is not a Student or one could
paraphrase it as Each Student is not a Student for some time, among the possible
options.
For temporal relationships, with the semantics r ∈ RI(t) → ∃t0 6= t.r ∈ /
I(t0 )
R , we devised two options that amounted to largely just shuffling around the
constituents of the template, with Ci an entity type and Ri a relationship:
(a) The objects in the facts in ..C1 .. ..R1 .. ..C2 .. do, at some time, not relate through ..R1 ..
(b) The objects participating in a fact in ..C1 .. ..R1 .. ..C2 .. do not relate through ..R1 .. at
some time.
Note that for an ontology setting, one can simply drop the C1 and C2 variables for
the entity types: The objects in ..R1 .. do, at some time, not relate through ..R1 .. .
The semantics of a basic temporal attribute is o ∈ C I(t) ∧ ho, di ∈ AI(t) →
0
0
/ AI(t ) , where the templates assume that any ‘has’ from the at-
∃t 6= t.ho, di ∈
tribute’s name is dropped if it is already in the name (e.g., hasColour), or it is
substituted with the verb in the attribute’s name. For instance, its option (e) with
an example about bonus payments would generate a sentence alike An Employee
receives a Bonus, but not always.

3.2. Dynamic constraints

The two basic types of dynamic constraints are extension and evolution (recall
Figure 2), where extension has the element remain a member of the source class
and in an evolution it stops being member of the source class once it evolves to
the target class. This may be at ‘some’ time or at a metric (quantitative) time,
and it may persist or not. They can be optional or mandatory, i.e., whether some
object, relation, or attribute may evolve or all objects/relations that instantiate
the class/relationship must evolve. This can be verbalised more or less harshly,
just like with atemporal constraints2 . That is, for future tense one can use the
auxiliary verb ‘will’ versus the stronger-sounding ‘has to’ or ‘must’. Likewise, for
the past there is a similar type of difference between ‘was already/before/earlier,
but not now’ vs. ‘must have been’. For instance, mandatory dynamic evolution in
I(t) I(t0 )
the past (DevM− ), i.e., o ∈ DevM− C1 ,C2 → (o ∈ C1 I(t) → ∃t0 < t.o ∈ DevC1 ,C2 )
I(t)
where the semantics of Dev is o ∈ DevC1 ,C2 → (o ∈ C1 I(t) ∧ o ∈ / C2 I(t) ∧ o ∈
I(t+1) I(t+1)
C2 ∧o ∈
/ C1 ), with, e.g., Butterfly (to be filled in for C2 in the template)
and the Caterpillar (C1 ) it used to be, with the following possible templates:
(a) Each ..C1 .. must have been a(n) ..C2 .. , but is not a(n) ..C2 .. anymore.
(b) Each ..C1 .. was a(n) ..C2 .. before, but is not a(n) ..C2 .. now.
(c) If ..C1 .. , then ..C1 .. was a(n) ..C2 .. before, but is not a(n) ..C2 .. anymore.
Persistence (PDex/PDev) has the change holding at all times in the future,
which is built from whatever will be chosen from the possible templates and
appended by a phrase like , and this remains so. or , and this remains so indefinitely..
For quantitative extension and evolution, we need a specific number for
counting and, implicitly, some time unit to be able to construct, e.g., ‘after

2 e.g., ‘each Prof teaches at least one course’ vs. ‘each Prof must teach at least one course’.
at least 3 years’. The number is denoted with the variable D1 in the tem-
plate. For instance, mandatory quantitative extension in future (QexM), i.e.,
I(t) I(t+n)
o ∈ QexC1 ,C2 → (o ∈ C1 I(t) → ∃(t + n) > t.o ∈ QexC1 ,C2 ), e.g., all Students (to
be slotted in at C1 ) have to Volunteer (C2 ) in the second year (D1 ) of their study,
with template options:
(a) Each ..C1 .. will also become a(n) ..C2 .. after [at least/at most/exactly] ..D1 .. .
(b) If ..C1 .. for [at least/at most/exactly] ..D1 .. , then ..C1 .. becomes a(n) ..C2 .. as well.
Regarding word choice in template design for relationships, the transition
concerns overlapping or successive processes; thus verbs such as ‘precede’ and
‘follows’ are applicable. On their own, they do not distinguish between the two
cases of whether the relations may co-exist or if one occurs after the other. This
can be addressed by disambiguating both cases, or one of the two with the other
left implicit. We chose the latter option, and add it to the evolution cases rather
than the extension cases, as they are stricter constraints. Likewise, ‘sequentially’
and ‘successively’ are roughly synonyms and imply that earlier state has ended,
so the only core difference to test is whether that should be made explicit, e.g.,
with inclusion of terms such as ‘ending’ or ‘terminating’, or not. For instance,
for mandatory dynamic extension for relationships in the past (RDexM− ), i.e.,
I(t) 0
ho, o0 i ∈ RDexM− R1 ,R2 → (ho, o0 i ∈ R2 I(t) → ∃t0 < t.ho, o0 i ∈ RDexR1 ,R2 I(t ) ),
where the semantics of ‘just’ RDex is ho, o0 i ∈ RDexR1 ,R2 I(t) → (ho, o0 i ∈
0
R1 I(t) → ∃t0 > t.ho, o0 i ∈ R2 I(t ) ), option (a) was Each ..C1 .. ..R1 .. ..C2 .. is preceded
by ..C1 .. ..R2 .. ..C2 ... For instance, every passenger who boards a flight must have
had a check-in process before with its template option (a) then results in a some-
what clunky Each Passenger boards Flight is preceded by Passenger checksIn
Flight. Note that also here it is easy to see how this template can be adapted
for ontologies by dropping the C1 and C2 variables.
These time markers are more challenging with the more complex con-
straints. For instance, mandatory dynamic evolution for relationships in the past
(RDevM− ), with the typical example that any pair of humans who are divorced
were married before that. The full set of template options to choose from was:
(a) Each ..C1 .. ..R1 .. ..C2 .. is strictly preceded by ..C1 .. ..R2 .. ..C2 .. .
(b) Each ..C1 .. ..R1 .. ..C2 .. is preceded by ..C1 .. ..R2 .. ..C2 .. and they are not in that ..C1 ..
..R2 .. ..C2 .. relation anymore.
(c) If ..C1 .. ..R1 .. ..C2 .. , then it was preceded by ..C1 .. ..R2 .. ..C2 .. and they are not in that
..C1 .. ..R2 .. ..C2 .. relation now.
(d) Each ..C1 .. ..R1 .. ..C2 .. must have been preceded by ..C1 .. ..R2 .. ..C2 .. and they are then
not in that ..R2 .. relation anymore.
(e) If ..C1 .. ..R1 .. ..C2 .. , then it must have been preceded by ...C1 .. ..R2 .. ..C2 .. , but ..C1 ..
not ..R2 .. ..C2 .. anymore.
Two dynamic attribute constraints were included in the evaluation, which
were uncontentious, and therefore they are omitted here due to space limitations.

4. Evaluation

The templates designed for the temporal elements and constraints are evaluated
aiming to answer:
1. Does each proposed template for a natural language sentence capture the
semantics of the temporal constraint adequately?
2. Which sentence among the options is preferred?
The principal approach is to use a two-pronged survey: a small group of experts
in temporal logic, and a ‘mixed’ group of experts in related relevant fields that is
not temporal logic, which includes modellers, NLG experts, and logicians.

4.1. Materials and methods

A survey was designed with the temporal constraints and elements, and 1-7 tem-
plates for each. The participants were given a brief written explanation of the
logic and how to read a template, an example for each constraint, and the scope
of the evaluation. They were instructed to evaluate each template on whether
it would capture the meaning adequately, which could be answered with either
“yes”, “sort of” denoting borderline, or “no”, and which of the templates they
preferred, if any, among the set of templates for that constraint. They also were
allowed to comment on each template option.
The three temporal logic experts are remote colleagues and were recruited by
email through purposive sampling. The mixed group also was recruited through
purposive sampling, with as prerequisite that they have a good to excellent under-
standing of modelling, logic, or NLG. No personal information was recorded other
than whether they have English as their mother tongue language. The answers
were collected in a pre-formatted spreadsheet. A follow-up interview with one
expert was also conducted to obtain further qualitative feedback on the choices.
All materials and results are available at http://www.meteck.org/files/
CREOL17suppl.zip.

4.2. Results and discussion

We present first the results of the experts and then those of the mixed group.

4.2.1. Experts’ judgements
An overview of the quantitative results is included in Table 1. 41% of the tem-
plates were deemed to properly represent the semantics of the temporal con-
straint, though only 12 templates received the same score (either all “yes”, “sort
of”, or “no”) with most (78) receiving two the same verdicts, reaching 89% then.
Four templates of constraints were unanimously preferred: those for Mandatory
Quantitative Extension and for Evolution (QexM and QevM) and the two at-
tribute constraints (Freez ‘frozen attribute’ and AQev ‘Quantitative evolution,
attribute’). With majority voting, this increased to 13 constraints. This is re-
markable, because the experts know each other well and have collaborated. Even
the simple constraint of ‘temporary class’ (Tc) had three different verdicts for its
option (b) ..C1 .. is an entity type whose objects are, for some time in their existence,
not instances of ..C1 .., as did option (b) of dynamic evolution in the past (Dev− ).
Quantitative extension and evolution used the grammatically correct ‘for’ in
the sentence, but this was deemed wrong. Instead, the experts preferred ‘since’.
This may be due to either the Since operator in the logic, or it has a stronger
sense of time and fewer senses than the multiple-use ‘for’, or that neither of the
experts has English as mother tongue. A follow-up interview on this matter with
one of the experts (e3 in the data) revealed that at least for e3 it was not so much
Table 1. Aggregates for experts and the mixed group (rounded) on whether a template captures
the semantics adequately; n: number; avg.: average; sd: standard deviation; pct: percent.

Temporal logics group Mixed experts group Both
n avg. sd. pct. n avg sd. pct. n avg. sd. pct.
Yes 125 42 15 41% 324 65 19 64% 449 56 21 56%
Sort of 98 33 16 32% 130 26 19 26% 228 29 18 28%
No 80 27 9 26% 44 9 8 9% 124 16 12 15%
Total 303 498

about ‘for’ vs ‘since’ but the fine-grained distinction between “for an x amount
of time since x” (or: holding between x ago and now) vs. “x chronons ago” but it
may not hold some time between x and now. On closer inspection, the logic stated
the latter, while the former was intended, which is what caused the confusion.
Regarding ‘preceded by’ and ‘followed by’ in the templates (recall Section 3.2),
there were a few comments such as “I don’t like “followed” that can be understood
as “at the next time point””. Thus, only using strictly/immediately to indicate the
next/previous time point and therewith leaving implicit when this may not be the
case when it is not stated explicitly was found to be ambiguous. Put differently:
an explicit time marker was perceived to be needed also for the ‘some time’. We
updated the templates accordingly (see online supplementary material).
An important result to note for the preferred templates, is that they do not
follow the structure of the axiom, regardless whether it would be rendered in the
formal semantics or DLRU S syntax. That is, the preferred templates amount to a
mapping between an axiom type rather than the axiom’s structure. To illustrate:
I(t)
– Formal Semantics: o ∈ DevM− C1 ,C2 → (o ∈ C1 I(t) → ∃t0 < t.o ∈
I(t0 )
DevC1 ,C2 ): “For objects involved in a mandatory dynamic evolution in the
past, if o is an instance of C1 , then there exists a time t0 earlier than time
t such that at that earlier time t0 it evolved from C1 to C2 .”
– DLRU S shorthand notation: C1 v 3− DevC1 ,C2 : “Each C1 is a subclass of
some time in the past dynamically evolved from C1 to C2 ”
– DLRU S full notation: C1 v C1 u ¬C2 u ⊕ (¬C1 u C2 ): “Each C1 is a subclass
of C1 and not C2 and at the next moment (not C1 and C2 )”
versus the preferred rendering Each C1 was a C2 before, but is not a C2 now. Not
linking each axiom component to a fragment of the natural language sentence
deviates from common practices of verbalising ontologies and conceptual models
(e.g., [13,25,27]) even when they may be using additional grammar rules to make
the output sentence grammatically better [7,11,16]. It might suggest that tem-
plates with fewer words were preferred. This is not substantiated by the data,
however: 12 preferences were sole or shared shortest template in the sense of
number of words, 9 were sole or shared longest template, and 3 were neither.
The follow-up interview with expert e3 included a clarification on the afore-
mentioned for/since issue. Expert e3 also noted that if neither expert has English
as first language, this may have contributed to the low amount of unanimity.
It may seem disconcerting that experts each use their own terminology, but e3
did not see that as a real issue, “for there is the logic that does have the precise
meaning anyway” and thus “resolves any confusion that may arise from using
slightly different terminology” (paraphrased). In this light, it is not likely that
asking many more temporal logic experts will make the results converge. Lastly,
e3 suggested that the relative large difference in “yes” between e3 and e1 (29 vs
62 ‘yes’) may be due to attitude toward judging, in that e3 aimed for one and at
most two “yes” per constraint. However, this does not explain why e2 had only
34 “yes”, but where no such criterion could have been used.

4.2.2. Mixed experts’ assessment results
The aggregate result for the mixed experts are included in Table 1. The main inter-
esting outcome is the large differences with the temporal logic experts on whether
the templates capture the semantics adequately. The mixed experts deemed many
more sentences to be fully or borderline acceptable. This is in large part explain-
able in that the experts have a better grasp of the subtleties of the temporal con-
straints. A clear illustration of this is RDevM− ‘Dynamic evolution for relation-
ships, past, mandatory’ option (a), where e1 had commented “Absolutely false!”
with 2 “no” and 1 “sort of” from the experts, yet it had 3 “yes” and 2 “sort of”
from the mixed group, and Sr ‘Snapshot relationship’ option (a) received 3 “no”
from the experts, yet 2 “yes” and 3 “sort of” from the mixed group.
Further, the aggregate evens out the more strict grading by the single mixed
expert who has English as first language (p1 in the data), which may also con-
tribute to assessing precision of verbalising the constraints, and one of the two
logicians who is well-versed in modal logics and thus may grasp the finesses of
temporal logics better. The aforementioned for/since issue of Qex− ‘Quantitative
extension, past, optional’ and QexM− ‘Quantitative extension, past, mandatory’
was absent from the mixed group, who deemed it mostly acceptable.
Regarding preferences for a particular template, there was no unanimous pref-
erence on any template, 4 where 4 out of 5 agreed on a preference, and 17 by ma-
jority voting (3 out of 5). As to unanimity of verdicts, the mixed group answered
the same on 6 templates. There is a slight preference for shorter sentences: 13
preferences were sole or shared shortest template, 5 were sole or shared longest
template, and 4 were neither.
Few comments were made by the mixed participants. Recurring ones in-
clude NLG expert (p2)’s option that several templates were either vague (e.g.,
option (a) of Tc ‘Temporal class’), repetitive (PDex/PDev ‘Persistent exten-
sion/evolution’), or that it “does not sound natural enough” (e.g., Dev− ‘Dy-
namic evolution, past, optional’). Participant p1 provided most feedback on word
choices and preferences; e.g., instead of ..., but is not now in Qev− ‘Quantitative
evolution, past, optional’ and QevM− ‘Quantitative evolution, past, mandatory’,
rather ..., but is not a C2 now, a dislike for the use of sequentially in dynamic
evolution templates (which the experts marked down as well), and on ceasing in
the template suggesting instantaneous change “rather than stating so explicitly”.

5. Discussion

The outcome with the preferred templates is, to the best of our knowledge, the
first proposal for linking temporal knowledge or information to natural language.
The number of participants is relatively small, but a larger group will neither
resolve the low inter-annotator agreement of the temporal logic experts, nor the
comparatively high ‘yay-saying’ among the mixed expert group whose difference
would likely be exacerbated with, say, a group of 4th year student participants
that are less conversant in temporal constraints. It is nigh on impossible to exam-
ine whether a non-expert group would really have understood the logic, because it
would require participants to communicate the meaning in a different mode than
the natural language, and those alternative modes were precisely a problem that
the natural language approach aimed to address. rendering it difficult to discern
whether a lack of understanding may be due to the other representation, Sim-
plicity or potential difficulty of the templates, as measured by number of words,
did not show unequivocal preference for shorter templates (hence, sentences), es-
pecially among the experts. The data collected cannot explain this, and it thus
merits further investigation whether some temporal constraints indeed cannot be
simplified further without losing their meaning.
Looking back at Figure 1, the temporal entities and constraints are now ver-
balised as shown in Figure 1-C. One might pose that once implemented, it may
end up in too many sentences for a modeller to check. However, one could 1) fur-
ther prioritise constraints and 2) rely on an automated reasoner. Regarding pri-
oritisation, the mandatory, past, and quantitative constraints are more interesting
from an information systems point of view, for they can be easily implemented
in a database as integrity constraints. Optional past and future constraints may
be interesting for querying and database updates only. Regarding the latter: with
the logic foundation, one could specify just a few important temporal constraints
and let the rest be inferred by the reasoner thanks to the logical implications
[4,14,17]; e.g., if the relationship is temporal, then so are the participating classes,
and if a superclass is temporal then so are its subclasses.

6. Conclusions

Template selection of temporal constrains in temporal logics and logic-based tem-
poral conceptual modelling languages showed low expert inter-annotator unan-
imous agreement, although 89% with majority voting. The experts were more
strict (41%) on whether a template captured the semantic than the mixed group
(64%), noting that there were few unanimous preferences. Taking into account
judgement of semantics, indicated preferences, and comments, all 34 axiom types
now have a template for verbalisation. The templates map as a whole to an axiom
type rather than their constituents.
We have recently completed a modelling experiment showing that, on the
whole, the template-based natural language was preferred over other notations
(semantics, DL, diagram, coding-style) especially for the more complex con-
straints [15]. Therefore, we plan to create bi-directional mappings between the
logic and the (pseudo-)natural language and design a multi-modal modelling in-
terface. We also expect that it then will be easier to collect data to investigate
the effect natural language renderings on modelling further.

Acknowledgments The author would like to thank the participants for their valu-
able feedback and the reviewers for their suggestions.
References

[1] I. Androutsopoulos, G. Lampouras, and D. Galanis. Generating natural language de-
scriptions from owl ontologies: the naturalowl system. Journal of Artificial Intelligence
Research, 48:671–715, 2013.
[2] A. Artale, E. Franconi, F. Wolter, and M. Zakharyaschev. A temporal description logic
for reasoning about conceptual schemas and queries. In S. Flesca, S. Greco, N. Leone, and
G. Ianni, editors, Proceedings of the 8th Joint European Conference on Logics in Artificial
Intelligence (JELIA-02), volume 2424 of LNAI, pages 98–110. Springer Verlag, 2002.
[3] A. Artale, R. Kontchakov, C. Lutz, F. Wolter, and M. Zakharyaschev. Temporalising
tractable description logic. In Proc. of the 14th International Symposium on Temporal
Representation and Reasoning (TIME-07), 2007. Alicante, June 2007.
[4] A. Artale, C. Parent, and S. Spaccapietra. Evolving objects in temporal information
systems. Annals of Mathematics and Artificial Intelligence, 50(1-2):5–38, 2007.
[5] N. Bouayad-Agha, G. Casamayor, and L. Wanner. Natural language generation in the
context of the semantic web. Semantic Web Journal, 5(6):493–513, 2014.
[6] H. Burden and R. Heldal. Natural language generation from class diagrams. In
S. Weissleder, L. Lúcio, H. Cichos, and F. Fondement, editors, Proceedings of the 8th In-
ternational Workshop on Model-Driven Engineering, Verification and Validation (MoD-
eVVa’11). ACM Proceedings, 2011. 17 October 2011, Wellington, New Zealand.
[7] N. E. Fuchs, K. Kaljurand, and T. Kuhn. Discourse Representation Structures for ACE
6.6. Technical Report ifi-2010.0010, Dept of Informatics, University of Zurich, Switzerland,
2010.
[8] D. Gianni, P. Bocciarelli, and A. D’Ambrogio. Temporal capabilities in support of con-
ceptual process modeling using object-role modeling. In Proceedings of the Symposium
on Theory of Modeling and Simulation-DEVS Integrative, 2014. Tampa, ?Florida, USA,
April, 2014.
[9] J. M. Giménez-Garcı́a, A. Zimmermann, and P. Maret. Ndfluents: An ontology for an-
notated statements with inference preservation. In E. Blomqvist et al., editors, Proc. of
ESWC’17, volume 10249 of LNCS, pages 538–654. Springer, 2017. 30 May - 1 June 2017,
Portoroz, Slovenia.
[10] H. Gregersen and C. S. Jensen. Temporal entity-relationship models - a survey. IEEE
Trans. Knowl. Data Eng., 11(3):464–497, 1999.
[11] N. Gruzitis and G. Barzdins. Towards a more natural multilingual controlled language
interface to OWL. In Proceedings of the Ninth International Conference on Computa-
tional Semantics, IWCS ’11, pages 335–339, Stroudsburg, PA, USA, 2011. Association for
Computational Linguistics.
[12] T. Halpin. Temporal modeling and ORM. In R. Meersman, Z. Tari, and P. Herrero.,
editors, OTM 2008 Workshops, volume 5333 of LNCS, pages 688–698. Springer, 2008.
[13] T. Halpin and T. Morgan. Information modeling and relational databases. Morgan Kauf-
mann, 2nd edition, 2008.
[14] C. M. Keet and A. Artale. A basic characterization of relation migration. In R. Meersman
et al., editors, OTM Workshops, 6th International Workshop on Fact-Oriented Modeling
(ORM’10), volume 6428 of LNCS, pages 484–493. Springer, 2010. October 27-29, 2010,
Hersonissou, Crete, Greece.
[15] C. M. Keet and S. Berman. Determining the preferred representation of temporal con-
straints in conceptual models. In 36th International Conference on Conceptual Modeling
(ER’17), LNCS, page (in print). Springer, 2017. 6-9 Nov 2017, Valencia, Spain.
[16] C. M. Keet and L. Khumalo. Toward a knowledge-to-text controlled natural language of
isiZulu. Language Resources and Evaluation, 51(1):131–157, 2017.
[17] C. M. Keet and E. A. N. Ongoma. Temporal attributes: their status and subsumption.
In H. Köhler and M. Saeki, editors, Asia-Pacific Conference on Conceptual Modelling
(APCCM’15), volume 165 of CRPIT, pages 61–70. Conferences in Research and Practice
in Information Technology, 2015. 27-30 January, 2015, Sydney, Australia.
[18] V. Khatri, S. Ram, R. T. Snodgrass, and P. Terenziani. Capturing telic/atelic temporal
data semantics: Generalizing conventional conceptual models. Transactions on Knowledge
and Data Engineering, 26(3):528–548, 2014.
[19] N. Marı́n and D. Sánchez. On generating linguistic descriptions of time series. Fuzzy
SetsandSystems, 286:6–30, 2016.
[20] R. Nelken and N. Francez. Querying temporal databases using controlled natural language.
In Proceedings of the 18th conference on Computational linguistics-Volume 2, pages 1076–
1080. Association for Computational Linguistics, 2000.
[21] E. A. N. Ongoma. Formalising temporal attributes in temporal conceptual data models.
Msc thesis, Department of Computer Science, 2015.
[22] N. Papadakis, K. Stravoskoufos, E. Baratis, E. G. Petrakis, and D. Plexousakis. PROTON:
A Prolog Reasoner for Temporal ONtologies in OWL. Expert Systems with Applications,
38:14660–14667, 2011.
[23] C. Parent, S. Spaccapietra, and E. Zimányi. Conceptual modeling for traditional and
spatio-temporal applications—the MADS approach. Berlin Heidelberg: Springer Verlag,
2006.
[24] A. Preventis, P. Marki, E. G. Petrakis, and S. Batsakis. Chronos: A tool for handling
temporal ontologies in protégé. In Proc. of ICTAI’12, 2012.
[25] H. Safwat and B. Davis. CNLs for the semantic web: a state of the art. Language Resources
& Evaluation, 51(1):191–220, 2017.
[26] T. Shunmugam. Adoption of a visual model for temporal database representation. M. IT
thesis, Department of Computer Science, University of Cape Town, South Africa, 2016.
[27] A. Third, S. Williams, and R. Power. OWL to English: a tool for generating organised
easily-navigated hypertexts from ontologies. poster/demo paper, Open Unversity UK,
2011. 10th International Semantic Web Conference (ISWC’11), 23-27 Oct 2011, Bonn,
Germany.