An analytical approach to building a core ontology for food
| Pages | 123-144 |
| DOI | https://doi.org/10.1108/JD-02-2016-0015 |
| Published date | 09 January 2017 |
| Date | 09 January 2017 |
| Author | Devika P. Madalli,Usashi Chatterjee,Biswanath Dutta |
| Subject Matter | Library & information science,Records management & preservation,Document management,Classification & cataloguing,Information behaviour & retrieval,Collection building & management,Scholarly communications/publishing,Information & knowledge management,Information management & governance,Information management,Information & communications technology,Internet |
An analytical approach to
building a core ontology for food
Devika P. Madalli, Usashi Chatterjee and Biswanath Dutta
Documentation Research and Training Centre,
Indian Statistical Institute, Bangalore, India
Abstract
Purpose –The purpose of this paper is to demonstrate the construction of a core ontology for food.
To construct the core ontology, the authors propose here an approach called, yet another methodology for
ontology plus (YAMO+). The goal is to exhibit the construction of a core ontology for a domain, which can be
further extended and converted into application ontologies.
Design/methodology/approach –To motivate the construction of the core ontology for food, the authors
have first articulated a set of application scenarios. The idea is that the constructed core ontology can be used
to build application-specific ontologies for those scenarios. As part of the developmental approach to core
ontology, the authors have proposed a methodology called YAMO+. It is designed following the theory of
analytico-synthetic classification. YAMO+is generic in nature and can be applied to build core ontologies for
any domain.
Findings –Construction of a core ontology needs a thorough understanding of the domain and domain
requirements. There are various challenges involved in constructing a core ontology as discussed in this
paper. The proposed approach has proven to be sturdy enough to face the challenges that the construction of
a core ontology poses. It is observed that core ontology is amenable to conversion to an application ontology.
Practical implications –The constructed core ontology for domain food can be readily used for developing
application ontologies related to food. The proposed methodology YAMO+can be applied to build core
ontologies for any domain.
Originality/value –As perthe knowledge, theproposed approachis the first attemptbased on the study ofthe
state of the artliterature, in terms of,a formal approach to the designof a core ontology. Also, theconstructed
core ontologyfor food is the first one as there is no such ontology availableon the web for domain food.
Keywords Methodology, Food, Core ontology, Application ontology, Knowledge discovery,
Knowledge mapping
Paper type Research paper
1. Introduction
Food is any substance of plant or animal origin containing essential nutrients, such as
carbohydrates, fats, proteins, vitamins, or minerals, and is consumed by living beings to
gain energy for survival and sustenance (Encyclopedia Britannica, n.d.). People are involved
in several streams of profession pertaining to food industry. Professions vary from a food
scientist to fitness experts and cooks. Each of the professions is in great demand and active
consequent upon which several application ontologies (where an application ontology is an
ontology, which is constructed following an application requirement of a domain; Dutta
et al., 2013) are built to improve search, navigat ion, permit inferences, improve
interoperability, which successively help the professionals in decision making.
Since an application ontology is constructed keeping in view the application requirement
of a domain (where a domain is any area of knowledge or field of study that we are
interested in or that we are communicating about that deals with specific kinds of entities;
Giunchiglia and Dutta, 2011), the performance of that ontology is usually undebatable.
Nevertheless, the major concern of the ontology community is the reuse of the ontologies.
It is observed that hardly any application ontology can be reused to build another
application of the same domain. If the application need varies slightly, in the majority of the
cases we end up with creating a new ontology from scratch. But we know, the creation of an
ontology is expensive, especially, in terms of time, infrastructural cost and human resources
(Farquhar et al., 1997).
Journal of Documentation
Vol. 73 No. 1, 2017
pp. 123-144
© Emerald PublishingLimited
0022-0418
DOI 10.1108/JD-02-2016-0015
Received 13 February 2016
Revised 17 August 2016
Accepted 18 August 2016
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0022-0418.htm
123
Core ontology
for food
Our goal is to reduce the cost of creation of the application ontologies. We hypothesize
that this is possible through a core ontology, where, a core ontology is an ontology that
includes the key concepts and properties required to describe a domain. Our idea is that we
create a core ontology for a domain and then reuse it by extending it further as and when
the application need arises. The identified concepts and properties belonging to the core
ontology are independent of any particular context-specific environment. The key
motivation of the current work is that even if a core ontology comprises a small set of
concepts, it would save a lot of time and expense of the domain developers interested in
creating application ontologies.
The current work aims to build a core ontology for the domain food, which can be reused
and thereby can be extended to application-specific ontologies for different contexts, for
instance online recipe search, dietary recommendation and food retailer. The constructed
core ontology is a self-sufficient one and would be able to answer the basic queries related to
food, for instance:
(1) What are the ingredients present in dish X?
(2) How many calories are present in dish Y?
(3) What are the dessert items available for a diabetic patient?
(4) Is this dish Z very spicy?
(5) What are the vegan dishes available for lunch?
The key challenge in the development of a core ontology lies in identifying and selecting the
key domain concepts and properties. Incidentally, as it has been discussed in the related
work (Section 2), there seems to be hardly any work on this topic. In this paper, we have
introduced a novel approach called yet another methodology for ontology plus (YAMO+),
for the development of a core ontology. The methodology is generic and domain
independent, and can be used to build core ontologies for any domain, for instance, music,
movie, real estate, architecture and tourism.
The rest of the paper is organized as follows: Section 2 discusses the current state of the
art, in general on core ontology and on food ontology; Section 3 discusses four use-case
scenarios related to food; Section 4 elaborates the development of the core food ontology
following the proposed methodology YAMO+; Section 5 provides some statistics of the
constructed core ontology; Section 6 concludes the paper providing directions to the future
research work.
2. Related work
Several attempts have been made to develop core ontologies in various fields. For instance,
Turki et al. (2014) have proposed a core ontology which helps to analyse organization’s
processes and refine their modelling. For requirement engineering (RE), new core ontology
which is proposed by Jureta et al. (2009) can determine whether RE-related issues can be
successfully tackled. It includes all the basic classes that stakeholders are concerned about
like beliefs, desires, intentions and evaluations. Hoekstra et al. (2007) describe legal core
ontology for legal knowledge systems. It aims to enhance knowledge exchange between
existing legal knowledge systems. The core ontology for legal knowledge interchange
format is designed based on two approaches proposed by Gruber (1993) and Uschold and
Gruninger (1996). Hunter (2003) illustrates the need to assimilate information from a variety
of genres of multimedia content within digital libraries that led to the development of core
ontology, thereby intensifying the semantic interoperability of audio-visual resources.
The ontology is developed using ABC model (Lagoze and Hunter, 2006). Audio-visual
description core ontology is designed by Issac and Troncy (2004) to cater to a wide range of
124
JDOC
73,1
To continue reading
Request your trial