=Paper=
{{Paper
|id=Vol-2137/paper_14.pdf
|storemode=property
|title=Classifying Processes and Basic Formal Ontology
|pdfUrl=https://ceur-ws.org/Vol-2137/paper_14.pdf
|volume=Vol-2137
|authors=Mustafa Jarrar,Werner Ceusters
|dblpUrl=https://dblp.org/rec/conf/icbo/JarrarC17
}}
==Classifying Processes and Basic Formal Ontology==
https://ceur-ws.org/Vol-2137/paper_14.pdf
Classifying Processes and Basic Formal Ontology
Mustafa Jarrar 1,2 and Werner Ceusters 1
1
Department of Biomedical Informatics, University at Buffalo, 77 Goodell Street, Buffalo, USA
2 Department of Computer Science, Birzeit University, Ramallah, Palestine
ABSTRACT functionings that are occurrents and those that are
Unlike what is the case for physical entities and other types of continuants. For example, intelligence and personality are
continuants, few process ontologies exist. This is not only because
processes received less attention in the research community, but also dispositions while behaviors and mental processes are
because classifying them is challenging. Moreover, upper level categories bodily processes. The memory image I have about my dad
or classification criteria to help in modelling and integrating lower level now (a cognitive representation) is related to my
process ontologies have thus far not been developed or widely adopted. remembering him now (a mental process). There are about
This paper proposes a basis for further classifying processes in the Basic
Formal Ontology. The work is inspired by the aspectual characteristics of
80 mental processes (e.g., learning, thinking, wanting,
verbs such as homeomericity, cumulativity, telicity, atomicity, arousal, and perception) and about 500 behavioral processes
instantaneity and durativity. But whereas these characteristics have been (e.g., cognitive, rhythmic and social behaviors). A related
proposed by linguists and philosophers of language from a linguistic ontology, the Emotion Ontology (Hastings et al., 2011)
perspective with a focus on how matters are described, our focus is on
distinguishes between three notions related to specifying
what is the case in reality thus providing an ontological perspective. This
was achieved by first investigating the applicability of these characteristics emotions: emotional processes, emotional dispositions, and
to the top-level processes in the Gene Ontology, and then, where possible, mental representations. About 170 types of processes have
deriving from the linguistic perspective relationships that are faithful to been covered in this ontology including emotion processes,
the ontological principles adhered to by the Basic Formal Ontology. mood processes, and emotional behaviors.
In BFO (Arp et. al. 2015, Smith et. al. 2012), processes
1 INTRODUCTION
and process boundaries are defined under occurrents, but
Living, growing, learning, purchasing, producing, sleeping, they are not elaborated further. Processes are those entities
and mating are examples of different types of processes, i.e. that occur, happen, unfold, or develop in time, have
entities with temporal parts that depend on other entities to occur. temporal proper parts, and depend on some continuant entity
The importance of process ontologies is rapidly to happen. Process boundaries are other types of occurrents
increasing in several domains such as in event discovery that occupy zero-dimensional temporal regions, thus they do
(Nevatia et al 2004, Li et al 2017), industry and engineering not have temporal parts and are not processes themselves.
(Morbach at 2007), software engineering (Ruy et al 2015), Examples of process boundaries are the moment of a
affective computing (Li et al 2016), among others. In person’s birth, and that what is described by terms such as
biomedicine, there are several ontologies that have been ‘midnight’, ‘departure’, and ‘arrival’.
recently developed or extended to cover a wide spectrum of In DOLCE (Masolo 2002), the corresponding entities are
types of processes, such as the process components of the called perdurants, which are those entities that happen in
Gene Ontology, the Emotion Ontology, and the Mental time, by accumulating different temporal parts: thus at any
Functioning Ontology, to name a few. The Gene ontology time t at which they exist, only their temporal parts at t are
(GO 2001) groups all processes under biological process, present. DOLCE was inspired for its classification by the
which is informally defined as a collection of molecular lexical semantics literature, thereby leaning on properties
events, specifically pertinent to the functioning of living such as homeomericity and cumulativity, which were used
beings and with defined temporal beginning and end. There to classify perdurants into states and processes if they are
are about 26k biological processes in the Gene ontology accumulative, or into accomplishments and achievements if
classified under 27 top-level processes, such as growth, they are non-accumulative. States are distinguished from
development, behavior, response to stimulus, metabolic and processes if they are homeomeric, and achievements are
immune system processes. These top-level processes were distinguished from accomplishments if they are atomic.
developed and are being extended without a formal or
Although DOLCE and BFO are based on distinct
foundational framework. The Mental Functioning Ontology
perspectives, it is not unreasonable to view processes and
(Hastings et al. 2012) was developed as a framework to
accomplishments in DOLCE as what are processes in BFO.
enable developing other ontologies related to mental health
However, how achievements and states are to be interpreted
and diseases. It distinguishes between the aspects of mental
by BFO remains unclear. If achievements are instantaneous
* To whom correspondence should be addressed: mjarrar@birzeit.edu
happenings with zero-time duration then they correspond to
1
Jarrar & Ceusters
BFO’s process boundaries. If, however, instantaneity is 2 GENE ONTOLOGY PROCESSES
meant to be a short period of time, then they are BFO To test the applicability of these notions, we attempted
processes. Additionally, states in DOLCE cannot be BFO to use them to analyze and annotate the top level processes
processes as they do not involve change. in the Gene Ontology. Table 1 illustrates a sample of these
Early research in philosophy and linguistics provided GO processes. The full annotations (and documentation of
different accounts of the so-called events, suggesting various choices) can be accessed online 1. We selected the 35 most
views and criteria to distinguish between actions, activities, top-level processes, that top all other (~45k) processes in the
accomplishments, achievements, processes, performances Gene Ontology.
states, mental and physical events, bodily movements, and It is worth noting that the top-level processes in the
others (c.f., Casati et al 2015). Most of this research was Gene ontology include not only process universals, but also
originally devoted to the semantics of verbal phrases, the so- defined classes, such as Single-Organism Process and
called ‘lexical aspect’ of verbs, and how they are structured Multi-Organism Process. Such process types cannot be
in relation to time. Figure 1 reflects our understanding of how classified according to criteria such as whether they are e.g.,
some of these accounts (e.g., Moens et al 1988, Bhatt 2005, homeomeric or telic because they are collections of different
Levin 2009) may be combined. (Vendler 1957) classified types of processes, defined based on the number of the
events into: activities, states, accomplishments, and organisms involved. Therefore, instead of selecting these
achievements. This proposal was later revised and extended processes, we selected and annotated their subtypes.
(e.g., Mourelatos 1978, Bach 1986, Krifka 1998, Caudal et
al. 2005, Trypuz et al 2007). Krifka made it very clear that This initial effort turned out to be a challenging task for
caution is required: these classifications are about predicates, two reasons: (1) shortcomings in the literature about what
i.e. descriptions, denoting entities such as processes in precisely is to be understood by these notions, and (2) their
reality, and are not classifications about processes focus on how matters are described (the linguistic
themselves: the same process can be described by distinct perspective), rather than on what is the case (the ontological
predicates each one of which can be classified differently. perspective).
Moreover, processes and events might be lexicalized in
different ways in natural languages. For example, the exact
same event instance of John’s speaking to Mary yesterday,
can be said in different ways, such as ‘I heard John speaking
with Mary, ‘John speaks with Mary’, or ‘John spoke with
Mary’. The difference between those English phrases is
ontologically irrelevant as they all refer to the same event.
Furthermore, there are verbs that do not denote or refer to
events, such as ‘It costs 20$’, ‘It weights 20kg’, or ‘It looks
easy’, which do not imply changes, as discussed earlier. Table 1. Sample of the top-level GO processes annotated.
Similarly, there are events that cannot be typically In the next sections, we examine the most commonly used
lexicalized using verbs, such as ‘war’ and ‘conference’. aspectual criteria and discuss whether and how they might be
In this paper, we examine the most commonly discussed used for further classifying processes in the spirit of BFO.
criteria in the literature to classify events and verbal
predicates: homeomericity, cumulativity, atomicity, telicity, 3 DEFINITIONS
durativity and instantaneity. Other aspectual notions, such as Central in our approach is the distinction between
incrementality and structure (Caudal 2005), distributivity temporal parts of processes, such as the first year of a
and collectivity (Champollion 2014), and quantization toddler’s life, and mere occurrent parts of processes, as
(Krifka 1998), are here not dealt with. each of the eating processes that are part of a toddler’s life. 2
p occurrent-part-of q (R1)
a primitive relation of parthood holding independently
of time between two process instances when one is a sub-
process of the other (Arp et al 2015:135).
1
http://github.com/mjarrar2/Processontology/wiki
2
From here on forward, we follow the standard typographical conventions
for particulars (lowercase, italics in definitions), universals (upper case)
and relationships (bold for particular-level relationships, italics for
Fig. 1. Overview of the aspectual classification of verbs. universal-level relationships) (Arp et. al 2015)
2
Classifying Processes and Basic Formal Ontology
P occurrent-part-of Q =def.: (R2) 3.1 Homeomericity
for every particular occurrent p, if p instance-of P, then
DOLCE considers an occurrence to be homeomeric iff
there is some particular occurrent q such that q instance-
all of its temporal parts, in BFO sense, are described by the
of Q and p occurrent-part-of q (Arp et al 2015:139).
very expression used for the whole occurrence, a view
which is close to the notion of homogeneity as defined in
p temporal-part-of q =def. (R3) (Dowty 1977:60). An example of a homeomeric process in
p occurrent-part-of q this DOLCE sense would be what is described by ‘sitting’.
& for some temporal region r p spans r Defined in this way, homeomericity is a property of
& for all occurrents c, r′ universals, not of instances. Although BFO is a
if (c spans r′ & r′ occurrent-part-of r classification of instances and not of universals, we can
then (c occurrent-part-of p iff c occurrent- work around this by recognizing several flavors of this
part-of q)) (Smith 2012, corrected). principle at the level of instances:
(1) p isotypic-part-of q =def. (R5)
The relation ‘p spans r’ in R3 is shorthand for ‘p p temporal-part-of q
occupies spatiotemporal region str which occupies temporal & p instance-of all types instantiated by q.
region r’. R3 allows us to define the corresponding Example (figure 2): p6 isotypic-part-of p5.
universal-level relation: (2) p weakly-homeomeric-in P =def. (R6)
P temporal-part-of Q =def.: (R4) all temporal parts of p which are not process boundaries
for every particular occurrent p, if p instance-of P, then are instances of P.
there is some particular occurrent q such that q instance- Examples (figure 2): p2 weakly-homeomeric-in DC1; p3
of Q and p temporal-part-of q. weakly-homeomeric-in DC1.
From R5 and R6 it follows that all parts of a process that
Figure 2 depicts an instance of PROCESS PROFILE, a is weakly-homeomeric-in some type are isotypic-part-of
subtype of PROCESS (Smith 2012). Displayed is a that process.
representation of an instance of a rhythm process profile, p1,
which could be an occurrent-part-of the beating of some (3) p strongly-homeomeric-in P =def. (R7)
person’s heart or of the performance of a piece of music by all temporal parts of p which are not process boundaries
a pianist and which in these cases would reflect the are instances of P and there is no such part of p that
variations in the number of heart beats produced by that instantiates a subtype of P.
person’s heart, resp. musical bars played by the pianist Example (figure 2): p5 strongly-homeomeric-in DC9,
(‘bpm’ = beats/bars per minute). The figure also depicts two p8 strongly-homeomeric-in DC11.
processes of equal duration, p2 and p3 which both are From R5 and R7 it follows that all parts of a process that
temporal-part of p1, as well as the further temporal parts is strongly-homeomeric-in some universal are isotypic-
p4, p5, p7 and p8 which also are each of the same duration, part-of that process.
be it half the duration of p2 and p3. Further depicted are the
Further to be investigated is whether it is worthwhile to
various defined classes instantiated by these processes.
introduce the DC ‘weakly-homeomeric-process’ as the DC
formed by all processes which are – or have an occurrent
part which is – weakly-homeomeric-in some process type.
Also whether it would pay off to implement the notion of
‘strongly-homeomeric-process’ as the DC formed by all
processes which are– or have an occurrent part which is –
strongly-homeomeric-in some process type or whether it
would be sufficient to have an axiom directly at the level of
the types to which it applies. In figure 2, this would hold for
DC8, 9, 10 and 11. It is also worth noticing that whether any
of the relations R5…7 holds, depends on whether (1) a pure
metaphysical stance is taken or (2) reality is looked at
through what an ontology allows us to see and what is
observable at the level of instances. If a restricted
ontological commitment would not allow us to observe or
describe instances p4…p8, then, under that commitment,
Fig. 2. Configurations of temporal parthood and homeomericity. e.g., p1 would be strongly-homeomeric-in DC5 and DC6.
3
Jarrar & Ceusters
3.2 Cumulativity Example (figure 2): DC5 cumulative-in DC1.
Cumulativity was extensively discussed in the lexical Clearly, if some process p is at least weakly-
semantics literature (see e.g., Krifka 1989, and Champollion homeomeric-in P then it is also instance-of a type which is
2014) in an attempt to describe verbs – not processes! – a cumulative-in P, but not vice versa. One could assume, for
distinction similar to what exists for mass nouns on the one example, that instances of Growth as defined in the Gene
hand and count nouns on the other hand: whereas two ontology would be weakly-homeomeric-in Growth and that
portions of water together make one (bigger) portion of water, all subtypes of Growth would be cumulative-in Growth.
two bottles of water together do not make one (bigger) bottle. Subtypes of Cell Aggregation, however, could be assumed
This view was later adopted in DOLCE for classifying to be cumulative-in Cell Aggregation, but for sure no
perdurants: a perdurant is cumulative if the mereological sum instance of it would be strongly-homeomeric-in Cell
of two instances of a type of perdurant is also an instance of Aggregation: not every temporal part of a cell aggregation
the same perdurant type; for example, ‘the sum of two sittings process is of the same type (e.g. its sub-process ‘cells
is still a sitting’ (Masolo 2002:24). Nevertheless, both this coming close to each other’ is not itself an aggregation).
definition and the example are rather unclear. However, the mereological sum of two cell aggregations
Looking at it from the BFO perspective, it leaves open would be a cell aggregation.
the question what sort of mereological sum is intended.
3.3 Telicity
Does it include occurrent parthood (see R1), as when my
sitting (p1) simultaneously with your sitting (p2) is summed From a lexical semantics perspective, an action predicate
to form p3 which is the sitting of the mereological sum of me is telic if it refers in one way or another to a terminal point
and you? Would certain process aggregates such as an for the action described as tending ‘towards a goal
orchestra playing a symphony being the mereological sum envisaged as realized in a perfective tense but contingent in
of the playings of the individual musicians plus the directing an imperfective tense’ (Garey 1957:106). An action
of the conductor? Or is it exclusively temporal parthood predicate is atelic if it does not mention any goal, purpose or
(R3) such as my sitting during the first 30 minutes of my endpoint, but can be used to denote an action as soon as it
total sitting and my sitting during the last 30 minutes? begins. For example, the predicate running is from such
The DOLCE documentation about cumulativity as well as perspective considered atelic because it does not mention
related proposals advanced in the linguistic community are not any goal or termination point. Furthermore, if somebody is
at all clear whether the sort of cumulativity they have in mind running a marathon (process p), then the predicate running
is a property of processes or of how a process is described. can be used to describe p immediately after its start, and
(Champollion 2014, Galton 2016), for instance, contend that independent of the outcome, i.e. whether or not the runner
cumulativity has more to do with the level of detail at which a finishes. The predicate running a mile is however
process is described, rather than what it is ontologically: if a considered telic (Krifka 1989:9): it mentions an endpoint
process p is described as ‘flying’ then cumulativity holds, as and it can only safely be applied to p after completion. The
the sum of any two parts of such a flying is also a flying. question we seek to answer is whether this notion of telicity
However, if the very same p is described as ‘flying from can in one or other form be applied to processes themselves,
London to New York’ then cumulativity, under their view, rather than to predicates under which these processes are
does not hold. From the BFO perspective, this meaning of described. In BFO terms: is telicity a notion that applies
cumulativity does not make sense: no entity becomes different only to representational units, or can it also be applied to
because it is described differently. As with homeomericity, that what the representational units are about? If the latter is
cumulativity as defined in DOLCE can be regarded as a the case, then telic processes would be those that have some
property of some universals which requires some work- terminal point or goal – whether or not these processes are
arounds in BFO, f.i. : described as such – and atelic ones those that do not.
p cumulative-with q =def. (R8) There are for sure processes that are telic under a
specific interpretation of ‘terminal point’, i.e. all those
all process types instantiated by p and all process types
which are finished! This is of course not what we have in
instantiated by q are instantiated by p, q and p+q.
mind here. What we do mean is that for a process to be telic,
Example (figure 2): p2 cumulative-with p3; DC1, DC5 there must be something ‘in it’ what we informally can
and DC6 are the all and only types that are instantiated by describe as ‘a change in the process’. However, this is just a
both p2 and p3, as well as p1 (the mereological sum of p2 and matter of speaking as under the perspective of BFO,
p3). processes do not change because they are changes!
P cumulative-in Q =def. (R9) Consider the current laws of physics on Earth, one being
P isa Q that objects such as a rock without support fall down to
& for all p1, p2 instance-of P: (p1+p2) instance-of Q.
4
Classifying Processes and Basic Formal Ontology
Earth. The rock cannot keep falling down forever, it be durative if it takes place over extended intervals, or
naturally stops when reaching a surface. Therefore, any such instantaneous (also called punctual) if it takes just a
‘falling under natural Earth conditions’ is a telic process. moment, i.e., a complete action with no explicit internal
This ‘falling’ terminates with (is followed by) a process that temporal structure, such as arriving and departure. We argue
can be roughly described as ‘the coming to a stop’ and that all processes are durative, and that there are no
which can take the form of bouncing a bit (on a hard processes that happen instantly in zero-time. We distinguish
surface), or penetrating a soft surface (sand, for instance); it between e.g., ‘the arriving of a train’ as a process extending
is this coming to a stop which is that what informally can be over time, and ‘the arrival of a train’ as a process boundary.
described as ‘the change in the process’. As another Nevertheless, the instantaneous events discussed in the
example, imagine that by looking through the linguistics literature are likely not meant to be time points or
representational units in figure 2 we do not just see the process boundaries, but rather, a sort of processes. Thus, we
process profile p1 and the parts p2, p3, … thereof, but the revise the definition of instantaneous processes to be those
actual beating of somebody’s heart b1 within that timeframe, types of processes that are the peak moments of some longer
and the process parts thereof – process parts being processes processes. Processes like, knocking, hitting, departure,
in their own right – b2, b3, …, such that p1 occurrent-part- arrival, and dying are examples of processes that extend
of b1, p2 occurrent-part-of b2, and so forth. Then b2 did not over time, but they have temporal parts that we call peak
‘change’ when b4 terminated and b5 started, and neither did moments. For example, knocking the door starts when one
b1. b2 is the change! Remember also that from the BFO moves his hand towards the door, then continues by
perspective it does not make sense to refer to processes ‘at a touching and hitting the door for a short moment (the peak
time’. We believe that telicity is once more not some unary moment), and then releasing the hand. Similarly, the
property of an individual process, but rather something in arriving of the train may start when the first trolley of the
relation to a universal. So we define: train starts entering the station, and decreases its speed into
p is-telic-in R =def. (R10) zero (the peak moment), wait a bit, and then open its doors.
p instance-of P Notice that peak moments are critical parts. For
& there exists some process q instance-of Q and example, if a process started with its agent intending it to be
some process r instance-of R, such that a knocking process but is terminated before its peak moment
(1) q not instance-of P, (2) p not instance-of Q, (i.e., before touching the door) then it is not a knocking, but
(3) p precedes q, and if it is terminated directly after its peak then it is a knocking.
(4) p and q are temporal-part-of r. In fact, this shows that instantaneous processes cannot be
Example (figure 2): each of p4 , p5, p7 is-telic-in DC1. strongly-homeomeric-in some universal because their
temporal parts before and after their peak are not of the
P is-telic-in R =def. (R11)
same type. One may even slice a process into a higher
for all process p instance-of P, p is-telic-in R. granularity and ask at what time point a process is
For example: falling (i.e. under normal earth conditions) considered done, but we believe that this is irrelevant since
is-telic-in moving (p is the falling proper, q is the coming to the boundaries of instantaneous processes are typically fiat
a stop (bouncing, penetrating, …), r=p+q is the entire boundaries. For example, the exact starting and ending of
motion process). Note that in general language, both p and r knocking, and similarly a train’s arrival, are typically a
might be denoted by ‘falling’ Another example: ‘walking matter of fiat determination. Further to be investigated is
leg swing’ is-telic-in ‘walking’. whether an entity (continuant or occurrent) is always created
We do not follow (Hennig 2008:262-4) according to at a peak moment, e.g. a sound when knocking, a new
which ‘a telic process can be interrupted any time but it can individual when giving birth, a hole when drilling a wall,
be only complete when its goal is reached’. Thus, if a person and so on. For this reason, no formalization for
on a cross road intends to cross over and lifts up his left leg instantaneous processes has yet been decided upon.
to start walking but gets immediately run over by a car, the
movement that was initiated was a ‘leg swing’, though not a
3.5 Atomicity
‘walking leg swing’. Our view is however compatible with Atomicity was adopted by DOLCE to distinguish
(Krifka 1998) in that if a process is telic (in P), then there is between accomplishments (non-atomic) and achievements
no temporal part of it that is telic (in P). (atomic) (Masolo 2002:24) but its definition is unclear. A
common understanding of it in the event semantic literature
3.4 Instantaneity is that it is a one-step change-of-state, i.e., lacking any
Some researchers distinguish between processes that are internal sub events (c.f., Caudal et al. 2005, Krifka 1998).
instantaneous and others that are durative (e.g., Garey 1957, But also this understanding of atomicity is problematic.
Mourelatos 1978, Krifka 1998). They consider an event to Since every part of the temporal region occupied by a
5
Jarrar & Ceusters
process is occupied by a process which is part of the larger Galton, A. 2016. The Ontology of Time and Process. Third
process, atomicity depends on the granularity level and is Interdisciplinary School on Applied Ontology, Bozen-Bolzano, Italy
subject to one’s perspective. As pointed out in (Krifka Garey, H. B. 1957. Verbal Aspects in French. Language 33, 91-110.
1989:80): “The notion of different granularities can be
Gene Ontology Consortium. 2001. Creating the gene ontology resource:
applied in cases where an entity appears as atomic under
Design and implementation. Genome Res, 11:1425–1433.
one description and as composed of different entities under
another description. For example, an arrival of a train at a Hastings, J., Ceusters, W., Smith, B., Mulligan, K. 2011. Dispositions and
station may be considered as atomic event or as an event processes in the Emotion Ontology. Proceedings of ICBO 2011.
which is composed of subevents… atomicity depends on the Hastings, J., Smith, B., Ceusters, W., Jensen, M., Mulligan, K. 2012.
position we take towards the minimal part problem’. Representing mental functioning: Ontologies for mental health and
Therefore, we do not find it opportune (yet) to propose a disease. Proceedings of ICBO 2012.
formalization compatible with BFO. Hennig, B. 2008. Occurrents. In Munn, K., Smith, B. (Eds.) Applied
Ontology: An Introduction, chapter 12 (pp 255-284)
4 CONCLUSION AND FUTURE DIRECTIONS
Krifka, Manfred. 1998 The origins of telicity". In Susan Rothstein (ed.),
We have overviewed some known aspectual notions Events and Grammar. Dordrecht: Kluwer, 197-235.
used to classify verbal phrases and examined their reuse to
Krifka, Manfred. 1989. "Nominal reference, temporal constitution and
build process ontologies under BFO. We provided BFO-
quantification in event semantics". In Semantics and Contextual
compatible interpretations of homeomericity, cumulativity
Expressions 75-115. Dordrecht: Foris.
and telicity, discussed instantaneity, and explained that
atomicity does not seem to be beneficial in classifying Levin, B. 2009. Aspectual Approaches to Lexical Semantic Representation.
processes. We illustrated the use of these notions to analyze Course LSA 116 UC Berkeley
and annotate the top levels of the Gene Ontology processes. Li, M., Wang, D., Lu, Q., Long, Y. 2016. Event Based Emotion
We plan to extend this work in several directions, including Classification for News Articles. PACLIC 30, 153.
the full ontological analyses of the Gene Ontology Li, C., Bendersky, M., Garg, V., Ravi, S. 2017. Related Event Discovery.
processes, in order to detect possible modeling Proceedings of WSDM '17. ACM, New York, USA, 355-364.
inconsistencies, as, for example, would be the case when a
Masolo, C., Borgo, S., Gangemi, A., Guarino, N., Oltramari, A., Oltramari,
non-homeomeric process is declared a subtype of a
R., Horrocks, I. 2002. The WonderWeb library of foundational
homeomeric process, extending BFO process categories, as
ontologies and the DOLCE ontology. WonderWeb deliverable D18.
well as examining the use of these notions at a larger scale,
such as restructuring of the process types found in WordNet. Moens, M., & Steedman, M. 1988. Temporal ontology and temporal
reference. Computational linguistics, 14(2), 15-28.
ACKNOWLEDGEMENTS Morbach, J., Yang, A., Marquardt, W. 2007. OntoCAPE -A large-scale
Part of this research was conducted during Prof. Jarrar’s ontology for chemical process engineering. Engineering applications
visit to the Ontology Research Center at UB, which was of artificial intelligence, 20(2), 147-161.
funded by the Fulbright Visiting Scholars Program. This Mourelatos, A. P. 1978. Events, processes, and states. Linguistics and
work was supported in part by CTSA NIH 1 UL1 philosophy, 2(3), 415-434.
TR001412-01 from the National Institutes of Health. Nevatia, R., Hobbs, J., Bolles, B. 2004. An ontology for video event
representation. Proceedings of CVPRW'04. IEEE
REFERENCES Ruy, F. B., Falbo, R. A., Barcellos, M. P., Guizzardi, G., Quirino, G. K.
Arp R, Smith B, Spear AD. 2015. Building ontologies with Basic Formal 2015. An ISO-based software process ontology pattern language and
Ontology. MIT Press. its application for harmonizing standards. ACM SIGAPP.15(2), 27-40.
Bach, E. 1986. The algebra of events.Linguistics & philosophy, 9(1),5-16.
Smith B. 2012. Classifying processes: an essay in applied ontology. Ratio.
Bhatt, R., & Pancheva, R. 2005. The syntax and semantics of aspect. LSA December 1; 25(4): 463–488
Summer Institute handout, Cambridge, MA.
Smith, B., et al .2012. Basic Formal Ontology 2. Specification and User’s
Casati, R., Varzi, A. 2015. Events, The Stanford Encyclopedia of Guide, (version 3/6/2014).
Philosophy (Winter 2015 Edition), Edward N. Zalta (ed.)
Smith, B., & Varzi, A. C. 2000. Fiat and bona fide
Caudal, P., & Nicolas, D. 2005. Types of degrees and types of event boundaries. Philosophical and Phenomenological Research, 401-420.
structures. Event arguments: Foundations and applications, 277-300.
Trypuz, R., Vieu, L. 2007. An ontology of the aspectual classes of actions.
Champollion, L. 2014. Distributivity, collectivity and Proceedings of LogKCA-07. 393-409.
cumulativity. Wiley’s companion to semantics.
Vendler, Z. 1957. Verbs and times. The philosophical review, 143-160.
Dowty, D. 1977 Toward a Semantic Analysis of Verb Aspect and the
English 'Imperfective' Progressive.Linguistics&Philosophy, 1(1):45-77
6