=Paper= {{Paper |id=Vol-1312/ldop2014_paper4 |storemode=property |title=Ontology Refinement Using Implicit User Preferences: A case study in cultural tourism domain |pdfUrl=https://ceur-ws.org/Vol-1312/ldop2014_paper4.pdf |volume=Vol-1312 |dblpUrl=https://dblp.org/rec/conf/jist/NasingkunISS14 }} ==Ontology Refinement Using Implicit User Preferences: A case study in cultural tourism domain== https://ceur-ws.org/Vol-1312/ldop2014_paper4.pdf 
  Ontology Refinement Using Implicit User Preferences:
        A case study in cultural tourism domain

          Krich Nasingkun1,2,3, Mitsuru Ikeda1, Boontawee Suntisrivaraporn2,
                                and Thepchai Supnithi3

                       1 Japan Advance Institute of Science and Technology

                                {krich, ikeda}@jaist.ac.jp
        2 Sirinthorn International Institute of Technology, Thammasart University, Thailand

                                      sun@siit.tu.ac.th
                         3 Language and Semantic Technology Laboratory

     National Electronics and Computer Technology Center (NECTEC), Pathumthani, Thailand
                                   thepchai@nectec.or.th




        Abstract. Recommender systems employ static knowledge elicited from ex-
        perts, causing high cost of maintenance for making the knowledge up-to-date.
        The contribution of this paper is the proposed method to collect potential con-
        cepts from users, in order to assist experts or development of automatic ap-
        proaches to refining an ontology. Implicit knowledge induced from the users,
        which is much less expensive to maintain ontology. Ultimately, it offers finer-
        grained, more effective recommendations that match expectation of the users.
        Keywords: Ontology refinement, implicit knowledge, cultural tourism




 1      Introduction

   Cultural tourism (or culture tourism) is a subset of tourism concerned with a coun-
try or region's culture, specifically the lifestyle of the people in those geographical
areas, the history of those people, their art, architecture, religion(s), and other ele-
ments that helped shape their way of life. The web site of Thai Cultural Knowledge
Center [1] is a cultural archive project, implemented through close cooperation be-
tween National Electronics and Computer Technology Center and Ministry of Culture
under the 2011 Memorandum of Understanding (MOU). The first phase of the project
was to develop a technology platform for acquisition, digitization, documentation,
preservation, security, and management of complex data in the cultural domain. The
second phase focused on integrating data from different sources using different stor-
age technologies, and providing a unified view of the collected data. From November
2010 to June 2013, the database contains more than 100,000 records, linking relevant
persons, organizations, places, and artifacts.
   It is quite difficult to find recommendations for tourists based on the cultural as-
pect, since there is abundant knowledge and data. Fig.1 shows an overview of the
recommender system framework for cultural tourism. The cultural portal is the central
database storing cultural data obtained by data collection module which is done by
officer in Ministry of Culture. To utilize the cultural database, an expert may con-
structs an ontology based on his/her expertise. Relation extraction is a key process for
eliciting knowledge in terms of ontology’s instances, concepts, and relations from
cultural database. Relation templates which are done in the ontology construction
process enable us to extract semantic relation among a focused set of entities in cul-
tural archive [2], which is constrained by relation types and their arguments. In this
paper, we focus on the ontology refinement process, to improve and clarify existing
knowledge. The better understanding provide the better alternatives for recommenda-
tion. User constrains (from user profile) and selection algorithm are deployed to cre-
ate the final recommendation output.




                   Fig. 1. Cultural tourism recommendation framework

   The rest of paper is organized as follows. Section 2 explains our related work.
Section 3 presents the proposed method and details of our algorithm. Section 4 illus-
trates usage scenarios of recommendation system that employ the proposed method.
Section 5 shows the discussion of this work. Section 6 provides conclusion and some
future development directions.
 2     Related Work

    Ontology refinement can be categorized into two approaches: semi-automatic and
automatic approaches. In the semi-automatic approach, the refinement algorithm aims
to help the knowledge engineer find relevant information. This can be done by nomi-
nating the terms to reduce the effort of looking for new relevant pieces of information.
An example of a technique could be the exploration of statistically significant terms.
Term co-occurrences are exploited to identify related terms based on statistical means
[3–4]. The automatic approach, on the other hand, does not require a knowledge engi-
neer during ontology refinement but require some principled way to drive the integra-
tion of new knowledge in the ontology. These automatic methods rely either on heu-
ristics (like some quality measure), or on information extraction from unstructured
source [5]; for example, the expansion of WordNet to the tourism domain [6]. In the
biomedical domain, an automated method to refine the Gene Ontology is proposed
[7]. The idea is to extract rules based on term variations for automatic term expansion
and validate them with the literature. By using IR techniques, the ontology query
model identifies missing knowledge in the ontology relevant to IR tasks. An automat-
ic method to revise the ontology accordingly is proposed for generating better queries
[8]. Many applied NLP techniques to this approach, but, to the best of our knowledge,
none of them concentrate on interests from system users. In our work, we focus on
semi-automatic technique to collect a potential concepts using evident from user in-
terest, in order to assist ontology engineer in culture domain.


 3     Ontology Refinement Framework

   Based on the definition, ontology refinement is a method to improve existing
knowledge to more clarify in specific domain. In our work, we proposed the ontology
refinement process based-on user interest, to collect the potential concepts which may
use to refine ontology in the future. Cultural tourism domain is used to demonstrate an
idea of our approach.


3.1    Resource Description Framework
   The Resource Description Framework (RDF) [9] is a framework for expressing in-
formation about resources. Resources can be anything, including documents, people,
physical objects, and abstract concepts. RDF is intended for situations in which in-
formation on the web needs to be processed by applications, rather than being only
displayed to people. RDF provides a common framework for expressing this infor-
mation so it can be exchange between applications without loss of meaning. Since it is
a common framework, application designers can leverage the availability of common
RDF parsers and processing tools. The ability to exchange information between dif-
ferent applications means that the information may be made available to applications
other than those for which it was originally created. RDF allows us to make state-
ments about resources. The format of these statements is simple. A statement always
has the following structure:

    

   An RDF statement expresses a relationship between two resources. The subject and
the object represent the two resources being related; the predicate represents the na-
ture of their relationship. The relationship is phrased in a directional way (from sub-
ject to object) and is called in RDF a property. Because RDF statements consist of
three elements they are called triples. Fig.2 show examples of RDF triples (informally
expressed in pseudo code).

        .
        .
        .
        .
        .
        

                Fig. 2. Example RDF statement (Source: RDF 1.1 Primer N.d.)

In the example above, Bob is the subject of four triples, and the Mona Lisa is the sub-
ject of one and the object of two triples. This ability to have the same resource be in
the subject position of one triple and the object position of another makes it possible
to find connections between triples, which is an important part of RDF's power


3.2     Cultural Tourism Ontology
   Cultural tourist has their specific character, their interest not only limit to target
destinations/activities itself. But they may gain knowledge in some more aspect
around cultural resources. Existing ontology-based recommendation approach has a
deep investigate on “is a” and “part of” relations, meanwhile the similarity measure-
ment among instances of the same or similar concepts are well investigated. As
shown in Fig.3, tourist make interest in “Wat Chong Kham” and “Grand Palace”.
Using “is a” and “part of” from existing ontology approaches, recommender system
may recommend another temple or palaces that related to user favorites. Limitation of
existing approached cannot capture interest that may related to resources in other
aspect. For example, “King”, “Art”, “Minority”, “Religious” or “Traditional and Ritu-
al” will never been concern.
                       Fig. 3. Conceptual Idea of Cultural Tourist Recommendation

   Table.1 show an example relations that we used to model cultural ontology in our
approached. Relations are defined to capture cultural aspects of cultural resources. By
this approached, cultural aspect of user interest will be analyses in order to recom-
mend the most related on some aspect.

Table1. Example Relations in Cultural Ontology

Relations              Domains                   Ranges                Meaning
has_periods (X, Y)     Festival                  Date/Time, Date,      X held on specific date or
                                                 Periods               period Y
has_location (X, Y)    Festival, Tradition       Location              X held on specific loca-
                                                                       tion Y
has_activities (X,     Festival, Tradition       Activities            X which includes the
Y1,Y2,..Yn)                                                            activities Y1,Y2,..Yn
to_celebrate (X, Y)    Festival, Tradition       Religious_event,      X has a purpose to cele-
                                                 Seasoning_event,      brate an Event Y
                                                 Living_activity
to_respect (X, Y)      Festival, Tradition       Animal, God,          X has a purpose to show
                                                 Water, Rice, etc...   respect/thanks to Y
sign_of (X, Y)         Event                     Religious             Event X is a sign of reli-
                                                                       gious Y
place_of_religious     Religious_attraction      Religious             X is a religious place
(X, Y)                                                                 belong to religious Y
founded_by (X,Y)       Attraction                Person, Organiza-     X is an important person
                                                 tion                  who support to create Y
lived_by (X,Y)         Attraction                Person, Organiza-     X is live in Y
                                                 tion
      1:   .
      2:   .
      3:   .
      4:   .
      5:   .
      6:   .



                   Fig. 4. Example RDF Statements in Cultural Ontology

   Fig.4 show a partial of cultural ontology that we will use to explain algorithm in
next section. Relation “founded by” and “is a” are a key relations that we will use to
identify potential concepts related to user interested. The output of our approached
can collected to assist ontology engineer, to refine ontology according to real interest-
ed of users.


3.3     Ontology Refinement Process




                          Fig. 5. Ontology Refinement Process

   As showed in Fig.5, Domain ontology and user favorite resources are used as an
input of Relation Analysis Process. By using evident from user favorite, the related
concepts and instances of user favorites are analyzed. Potential concepts will be nom-
inate by Potential Concept Analysis Process, all relations of related concepts are take
into account. Possible concepts that may clarify user interest are formulated. Howev-
er, only the concepts that share common relation are nominated as a potential new
knowledge. In Personalized Ontology Process, new knowledge are collect in order to
assist expert to update existing domain ontology in future.
3.4     Ontology Refinement Algorithm

        In this section, we explain the pseudo code for ontology refinement. The input
of our framework is a set of users favorite’s resources that input directly from users.
Let O be an ontology that modeling by RDF statements, F = {f1, f2, f3… fn} be a set of
users’ favorite instances. First, we look at each instance to find identify domain con-
cept of relation DC and the related instances RI. Next, we create a temporary concept
that contain only a related instances in each relations, then list of concept that related
instances belong to set of relation concept RC. the intersection operation is used to
identify the unique relation that share common between the same domain concept and
range concept. Finally the set of relation Result which store the list of relation of con-
cepts are return as an output. The output of our approach is a personalized extended
ontology.

  Algorithm 1: Refinement Algorithm

  Input: O = {rdf triples}; F set of n user’s favorite individuals from O
  Output: {extended rdf triples} w.r.t O and F (to be added to the personalized ontology)
  1: ADC  ; Result  ;
  2: for each f  F do
  3:     DC  ; RI  ;TempC  ;RC  ;Relation  ;
  4:     DC  {c | f is an instance of concept c}
  5:     RI  {g | r (f, g) for some role r}
  6:     TempC  {G | G is a temporary concept that has only RI as a member}
  7:     RC  {G | r (f, g) for some role r that g is an instance of concept G}
  8:     Relation  {r (DC, RC) and r (DC, TempC) | r (f, g) for some role r}
  9:     if not exist (DC in ADC) then
  10:                ADC ADC  DC
  11:                 Result  Result  Relation
  12:    else
  13:                 Result  Result  Relation
  14: end if
  15: end for
  16: return Result


   The output from our technique capture the real interest concepts based-on existing
ontology structure in the user point of view. Ontology engineer can use this technique
to collect the realistic concepts to assist the ontology refinement process.


 4      Recommendation Scenarios

    This section shows the usage scenarios recommendation system in Cultural rec-
ommendation framework. Existing ontology has the structure as shown in Fig.4, tour-
ist identify “Wat Arun” as his favorite place. By our approach, all related instances
and concepts will be investigate. The unique character of user interest will be identify.
Finally, the potential knowledge are nominate as an extended of personalized ontolo-
gy.




          Fig. 6. Example Ontology Extension Generated by the Proposed Technique

    Fig.6 show an example user favorite instances and existing ontology structure. The
output of our approached is an extended knowledge, Concept “Temple founded by
King”, “Temple founded by Rama II” and “Rama II” are nominated to be the part of
the personalized ontology.
    In the recommendation, instead of interpret the interest to temple according to ini-
tial ontology. The emerging of new concept ‘Temple founded by King’ lead the inter-
pretation of user interest into a specific group of temple. Instead of 40,717 temples in
Thailand, we may scope the number of recommendation using our approach into 217
items. When we collect the list of concepts from many users, ontology engineer will
use it to analyze and refine existing domain ontology in the future work.

 5     Discussion

  In our approach, induced related concepts require instances as supporting evidence.
For example, if we have instance of temples as a member of user favorites, it is possi-
ble to discover sub concepts of the temples. The shared commons among different
types of concepts (ex. Festival, Palace, Temple) without instances, cannot infer the
new knowledge. Although we can identify the links among items, the support evi-
dence (instant of concept) still required to prove the intention from users. For exam-
ple, our technique cannot infer concept ‘King founded Temple’ from favorite instance
of Palace or Buddhism_Related_Festival concept (even we have some relation be-
tween this two concepts).
  Without this approached, existing taxonomy of ontology can produce the similar
outputs (for example Temple founded by King). However, that concept may not be
interested (never be used) by real users. In addition, it will make the over size of on-
tology problem. In contrast, expert will decide to approve/ignore the inferred
knowledge in our approached.

 6     Conclusion and Future work

  We have presented an ontology refinement approach in cultural tourism domain us-
ing implicit knowledge induction from users’ favorite resources. New potential con-
cepts based-on user interest are discover to improving and clarifying the existing
knowledge.
   Some future work includes an implementation of a recommendation framework for
cultural tourism and evaluation of the recommendation result. In addition, the ontolo-
gy refining approaches can be applied to other specific user-oriented domains.


Acknowledgment

   This research was conducted under a grant in the SIIT-JAIST-NECTEC Dual Doc-
toral Degree Program.


References

1.      Cultural Knowledge Center, Ministry of Culture, Thailand. http://www.m-
        culture.in.th, accessed September 15, 2014.

2.      Buranasing, W., Sornlertlamvanich, V., Charoenporn, T.: Semantic Relation
        Extraction for Extensive Service of a Cultural Database. Proceedings of the
        SNLP2013. pp. 241-247 (2013).

3.      Faatz, A., Steinmetz, R.: Ontology enrichment with texts from the WWW.
        Semantic Web Mining. 20 (2002).

4.      Cimiano, P., Handschuh, S., Staab, S.: Towards the self-annotating web. Pro-
        ceedings of the 13th international conference on World Wide Web. pp. 462–
        47 (2004).

5.      Hahn, U., Schnattinger, K.: Towards text knowledge engineering. Hypothesis.
        1, (1998).

6.      Navigli, R., Velardi, P.: Automatic Adaptation of WordNet to Domains. Pro-
        ceeding of the LREC2002. pp. 1023-1027 (2002).
7.   Lee, J.-B., Kim, J. -j., Park, J.C.: Automatic extension of Gene Ontology with
     flexible identification of candidate terms. Bioinformatics. 22, 665–670
     (2006).

8.   Jimeno-Yepes, A., Berlanga-Llavori, R., Rebholz-Schuhmann, D.: Ontology
     refinement for improved information retrieval. Information Processing &
     Management. 46, 426–435 (2010).

9.   RDF 1.1 Primer N.d. http://www.w3.org/TR/2014/NOTE-rdf11-primer-
     20140624/, accessed October 5, 2014.