=Paper=
{{Paper
|id=None
|storemode=property
|title=A Domain Ontology to Support Evidence-Based Practice and Context Usage on Crime Prevention
|pdfUrl=https://ceur-ws.org/Vol-776/ontobras-most2011_paper1.pdf
|volume=Vol-776
|dblpUrl=https://dblp.org/rec/conf/ontobras/LopesRVB11
}}
==A Domain Ontology to Support Evidence-Based Practice and Context Usage on Crime Prevention==
https://ceur-ws.org/Vol-776/ontobras-most2011_paper1.pdf
A Domain Ontology to Support Evidence-Based Practice
and Context Usage on Crime Prevention1
Expedito C. Lopes1, Gabriela P. R Pinto2, Vaninha Vieira3, Teresinha F. Burnham4
1
Program in Systems and Computing – Salvador University (UNIFACS)
Salvador – BA – Brazil
2
Informatic Area – Department of Exact Sciences – State University of Feira de Santana
(UEFS) – Feira de Santana - BA – Brazil
3
Computer Science Department – Federal University of Bahia (UFBA)
Salvador – BA – Brazil
4
Faculty of Education – Federal University of Bahia (UFBA)
Salvador - BA – Brazil
ditoexpe@gmail.com, vaninha@dcc.ufba.br, gabrielarprp@gmail.com,
tfroes@ufba.br
Abstract. Evidence-Based Practice (EBP) represents a decision-making
process centered on justifications of relevant information. Context is a kind of
knowledge that supports the ability to define what is or is not relevant in a
given situation. The decision-making context can have an impact on evidence-
based decision-making, but the integration of evidence and context is still an
open issue. Ontology is referred as the shared understanding about a domain.
One of the main reasons for developing context models based on ontologies is
the knowledge sharing that enables computational entities, such as agents and
services to find actors’ similar profiles in decision making environment. This
paper presents the integration of evidence and context on decision making and
proposes a domain ontology to support EBP and context usage on the crime
prevention domain. A practical implementation serves to validate our work.
1. Introduction
The Evidence-Based Practice (EBP) paradigm, usually employed in several areas such
as Medicine, Crime Prevention, Education and Software Engineering, are systematic
procedures that take into account a problem being faced by an actor (e.g. diabetes in
children), his/her needs and preferences for decision, leading to a search for evidence
and an application based on the best research evidence found (Sacket et al. 2001). The
procedures represent an evidence-based decision-making process, centered on
justifications of relevant information (Dobrow et al. 2004).
Context is a knowledge that supports the ability to define what is or is not
relevant in a given situation (Vieira et al. 2010). The application of evidence to a
1
This work was supported by the National Institute of Science and Technology for Software Engeneering
(INES+), founded by CNPq grant 573964/2008-4. http://www.ines.org.br.
13
particular patient, for example, detains important contextual information in the EBP
procedures and includes comparative analysis between different contexts: that of the
generation of evidence and that of the patient.
According to Dobrow et al. (2004) “the two fundamental components of an
evidence-based decision are evidence and context. The decision-making context can
have an impact on evidence-based decision-making”. There is significant research in the
fields of EBP and context. However, the integration of evidence and context in
computer models is still an open issue.
Ontology is referred as the shared understanding of some domains, which is
often conceived as a set of entities, relations, functions, axioms and instances. One main
reason for developing context models based on ontologies is knowledge sharing that
enables computational entities, such as agents (human or software) to find similar
profiles in decision making environment (Wang et al. 2004).
This paper aims at: (i) presenting the integration of evidence and context
concepts preserving their characteristics of representation for domains that use EBP; and
(ii) describing domain ontology to support the search and retrieval of evidence,
regarding their contexts. This ontology is a start point to provide arguments for a
semantic formulation about the characteristics of a problem, increasing the evidence-
based solution, in the crime prevention domain. The motivation behind the ontology
construction is due to the lack of ontologies adaptable to our purpose. Therefore, this
paper also aims at providing artifacts to support system designers and provenance
community experts.
The evidence retrieval increased with contextual information can also facilitate
reapplying decision-making justifications when similar problems occur. The context
usage also allows filtering out and sharing more useful information so the retrieved
information can meet the decision maker needs. In this sense, context becomes a
significant tool to optimize performance and reduce search results. Filtering
mechanisms avoid more explicit user interactions with the application (Bunningen,
2004).
The remaining of the paper has the following organization. The key concepts
regarding context and evidence are described in Section 2. Section 3 presents a meta-
model that integrates evidence and context high level concepts. In Section 4, the domain
ontology for the Crime Prevention domain, integrated with the meta-model concepts, is
described. An implementation for a scenario of usage is presented in Section 5, which
serves to validate our work. Related Works are described in Section 6. Finally, in
Section 7 we present our conclusions and directions for further work.
2. Background
This section defines context and provides an overview of Evidence-Based Practice.
2.1. Context
There are several definitions about context. A classical definition is proposed by Dey
and Abowd (2001) for whom context is “any information that characterizes the situation
of an entity, where this entity is a person, place or object considered relevant in the
14
interaction between the user and an application. A context is typically the location,
identity and status of people, groups and computational and physical objects”. Context
can also be seen as a set of conditions and relevant influences that make a situation
unique and understandable (Brézillon 2007) or as a set of information items (e.g.
concepts, rules and propositions) associated with an entity (Vieira et al. 2010).
An item is considered part of a context only if it is useful to support a problem
solving. This item corresponds to a contextual element defined as “any data, information
or knowledge that enables one to characterize an entity on a given domain” (Vieira et al.
2010). Contextual information regarding acquisition is: (i) given by the user, whether
from persistent data sources or from profiles; (ii) obtained from a knowledge base; (iii)
obtained by means of deriving mechanisms; or (iv) perceived from the environment
(Henricksen and Indulska 2006). It is usually identified through the dimensions why,
who, what, where, when and how (Brézillon 2007).
One step in the task execution or problem-solving process is known as focus.
The contextual elements should have a relevant relationship to the focus of a human
agent or software agent. In general, focus is what determines which contextual elements
should be instantiated (Brézillon 2007).
2.2. Evidence-Based Practice
According to Thomas and Pring (2007), in general, information labeled as evidence is
those whose collection has concerns about its validity, credibility and consistency with
other facts or evidences. In relation to its credibility, evidences are categorized in ways:
1. Based on professional practice, as a clinical examination;
2. Generated by a process involving scientific procedures with a proven history in
producing valid and reliable results, e.g a collect performed by biomedical;
3. Based from published research that corresponds to critical reviews of the area,
such as randomized clinical trial.
“Evidence” in EBP, also called “research evidence”, corresponds to the third
category above and means a superior type of scientific research proof, such as generated
through systematic review and meta-analysis in the highest level. These published
researches are available in reliable data bases, usually found on sites over the Internet,
carried out by independent research groups (Sackett et al. 2001). This is the concept of
evidence applied in this paper.
To clarify further, a systematic review is a review that presents meticulous
research and critical evaluations of primary studies (case study, cohort, case series, etc.),
based on research evidence related to a specific theme. It contains analysis of qualitative
results conducted in distinct locations and at different times. Meta-analysis is a
systematic review of qualitative and quantitative characteristics (Friedland et al. 1998).
Evidence-Based Practice (EBP) involves complex decision-making, based on
available research evidence and also on characteristics of the actor of the problem,
his/her situations and preferences (Sackett et al. 2001).
In the medical area, EBP focus is to provide effective counseling to help patients
with terminal or chronic illness to make decision in order to extend or increase the
quality of their life (Friedland et al. 1998). What is objectively searched is “the
15
integration of best research evidence, clinical skill and patient’s preferences, regarding
individual risks and the benefits of proposed interventions” (Sackett et al. 2001).
In crime prevention, EBP involves the correlations practice that has been proven
through scientific research, aimed at reducing the recidivism of offenders. EBP
primarily considers the risk and need principle of the offender, besides the motivation,
and treatment and responsibility principles (Warren 2007).
The EBP focus for education area is improving the quality of research and
evaluation on education programs and practices, and hence, the information diffusion in
the educational research field to be used by professionals and policies creators (Thomas
and Pring 2004).
So, we generalize the EBP steps in the following way:
1. Transforming the need for information into a question that can be answered;
2. Identifying the best evidence to answer the question;
3. Critically analyzing the evidence to answer:
• Is it valid (appropriate methodology and proximity to the truth)?
• Is it relevant (size and significance of the observed effects)?
• Can it help (applicable in professional practice)?
4. Integrating critical analysis with professional skills and the values and cultural
aspects of the actor of the problem answering:
• How much the evidence can help the actor in particular?
• Is it adaptable to actor’s goal and preferences?
• How much safety can be expected?
5. Evaluating the efficiency and effectiveness of the results of each step for future
improvement.
3. A Meta-model to Represent Evidence-Based Practice and Context Usage
The primary aim of a meta-model is to provide a set of building blocks and rules used to
build models (Chomsky 1965). In this perspective we propose a class structure that
represents information related to EBP procedures, while taking into consideration
information about its decision-making context.
Thus, domain analysis was done in the crime prevention (particularly in juridical
and social work), medical and educational environments, including: bibliographical
research (Warren 2007; Satterfield et al. 2009; Sacket et al. 2001; Friendland et al.
1998; Thomas and Pring, 2004, etc.), specific legislation research, analysis of real cases
collected and interviews with decision-makers from Pernambuco state court, Brazil.
Figure 1 below presents a meta-model that corresponds to integration of EBP
with contextual information. We use the extension construct stereotype of the UML to
select enumerated values. To facilitate its presentation in a systematic way, it became
convenient to group classes in two integrated packages: Context and Evidence.
3.1. Context Package
The classes of the context package are based on Vieira et al. (2010). The focus is treated
as an association of a task with an agent, which have a role in problem resolution. A
16
task "make a critical analysis of the best evidence found" for a “medical” agent in the
role "evaluator", serve as example. ContextualEntity represents the entities of the
application conceptual model and is characterized by at least one contextual element. A
contextual element is a property that can be identified by a set of attributes and
relationships associated with ContextualEntity (Vieira et al. 2010). Accessibility is an
example of a contextual element for the Document class. The association between Focus
and ContextualElement determines what is relevant for a focus.
Characteristics attributed to the type of context (dimension) and the method of
acquiring contextual elements are considered in the framework. Contextual sources may
be internal or external to the decision-making environment (e.g., the patient’s medical
records, a document with evidence obtained from websites).
Figure 1. A meta-model integrating evidence with context
3.2. Evidence Package
The starting point is the observation of a problem presented by an actor to be decided by
agent. Each problem is associated with an inquiry that is initiated by a formulated
question (see step 1 of the EBP procedures), and completed with a self-evaluation of the
research performance and suggestions for the future (see step 5 of the EBP procedures),
whose information is instantiated in the Research class. Each domain in which EBP is
applied has a list of different types of questions. For example: "diagnosis" and
"prognosis" in the medical area, "drug testing" and "occurring disorders" in the area of
crime prevention, and "educational research" in education.
During the evidence research, several searches can be performed to retrieve
documents. For the Seek class, the expression and the type of search must be present.
17
InformationSource represents database sources that hold documents with research
evidences, such as Cochrane Library, Campbell Library or Springer International
Publisher, and the evidences were generated by independent research groups of specific
areas (e.g. Cochrane Collaboration for medical area and Campbell Collaboration for
areas of education and crime prevention).
Each document presents a type of study that can be in all domains (e.g.
systematic review, case study) or in specific domain (cohort - in the medical area;
action-research - in education). Systematic review and meta-analysis are studies of
second degree; the remains are of first degree (Friedland et al. 2001).
After finding evidences, the agent (decision maker) will choose the one that
seems the most appropriate (step 2 of EBP), which is instantiated in the Evidence class.
The result of the critical analysis – the validity, relevance and applicability of the
best evidence (step 3 of EBP) – corresponds to contextual information. Relevance is a
contextual element in Document, while applicability (practical utility) is in Evidence.
Thus, Document and Evidence are specializations of ContextualEntity.
The Intervention class is the result of an association among the Problem, Actor
and Evidence classes. It contains a description of a decision made (intervening solution)
where information about associated classes have been considered including preferences,
values and cultural aspects (conduct, behaviour, for example) of the actor with the
problem presented (step 4 of EBP). A preference is a contextual element and hence
Actor is a specialization of ContextualEntity. Problem aspects, such as the
circumstances about a juridical fact for the criminal area, generally, are contextual
elements used to diagnose the problem. So, the Problem class is a ContextualEntity too.
Summarizing, some elements that characterize a meta-model are shown through
some examples. The Agent class corresponds, respectively, to the Doctor, Judge and
Professor classes for the medical, juridical and educational areas. Evidence and Seek, for
example, are general classes for any domains. The classes ClinicalProblem, in the
medical domain, and JuridicalFact, in the juridical domain, represent the Problem class
of the meta-model.
4. A Domain Ontology for the Crime Prevention
This section describes the main steps in the construction of ontology for representing
EBP considering contextual information in the crime prevention domain. The ontology
is constructed using the Web Ontology Language (OWL). To edit the ontology and
axioms we used the Protégé (http://protege.stanford.edu).
4.1. Ontology Concepts
Figures 2 and 3 show a set of subclass/superclass of the main concepts defined in the
crime prevention ontology. The concepts were constructed based on a survey of the
concepts related found in the technical and scientific literature (Warren 2007; Gomes
2008; Moreira 2007; Saliba 2009). In this section, we present the specific concepts for
the crime prevention domain, since the high level concepts concerning context and
evidence were described in the meta-model (Section 3).
18
The ontology comprises two main classes: ContextualEntity and
ConventionalEntity. ContextualEntity contains the subclasses that detain at least a
contextual element (or contextual property), which supports the description of scenarios
found in environmental decision making. According to the meta-model, described in
Section 3, ContextualEntity has six main subclasses: Agent, Actor, Problem, Document,
Evidence and Intervention. For the crime prevention domain, we defined the subclasses
illustrated in Figure 2 and summarized the main contextual properties in Table 1.
Figure 2. Subclasses of ContextualEntity for the crime prevention domain
Table 1. Descriptions of the main contextual properties - crime prevention domain
Subclass Contextual Description
Properties
Judge hasExpertAffinity identifies a relation of expertise from the Judge profile on a given subject
matter (e.g. crimes against children). It helps to identify mutual affinities
among judges optimizing performance and reducing search results
Participant hasAbilities represents the defendant’s or victim’s skills, and is used to find mutual
affinities with intervention programs (e.g. revenue). Reduce search results
Defendant hasPotentialRisk comes from juridical and psychosocial evaluations (profile). Behavior data,
conduct, fact description and given sentences, especially for recurrent cases,
are bases to characterize an offender’s degree of risk.
JuridicalFact hasCircumstances describes the information about time and geographic aspects for the occurred
fact. Information about number of people involved and their attitudes are
desirable too. It is relevant and determinant to understand the juridical fact
hasValidity indicates whether the document should be selected based on its quality and the
Researched methodological rigor associated with the question asked by the decision maker
Document (judges, in this case)
hasRelevance indicates whether the set of results (outcomes) in the document, often
presented in statistical form, is consistent and significant
Evidence hasApplicability indicates whether the evidence presented in the document is credible in the
context of other knowledge, or whether it has practical utility in general
hasAdaptability indicates the degree of coherence in the application of evidence for the
conducted behavior, needs and preferences of the defendant (or victim)
Restorative hasSafety denotes the percentage of safety that the judge have to apply the specific
Intervention evidence to a particular participant (defendant or victim)
hasExpectation refers to the percentage of support expected from the use of evidence in
relation to the participant (defendant or victim)
ConventionalEntity contains all the specific subclasses of the domain, which
does not have a direct influence of context. They are JuridicalResearch, Seek and
JuridicalEvidenceProvider as shown in Figure 3. For the JuridicalResearch subclass,
the property historic should include general comments and the number of documents
that were accepted and rejected, besides of the properties presented in Figure 1. In the
19
Seek subclass, properties about the researched document validity time must be
considered. They are specific to instantiate the start and the end of the document validity
found. The JuridicalEvidenceProvider subclass contains the follow properties: name,
that mean the name of the Internet site accessed, and homepage, that detain the URL
address.
Besides the enumerations presented in Figure 1, for the crime prevention
domain, we add a subclass BasicProgram with the following values instantiated: citizen,
revenue, education, psychosocial, psychiatric and shelter. This subclass serves to
support the conventional intervention (no evidence-based) that exists in the courts. All
subclass with enumerations are children of the TypeClass class.
Figure 3. Subclasses of ConventionalEntity for the crime prevention domain
The figures and table of this subsection were not fully developed in this article
due to the limits of space.
4.2. Inference Rule
Some inference rules have been built and others are in development. In the following
section addresses the applicability of some of them. For the sake of space we will not
describe the properties of all classes in this paper. We will mention briefly the
characteristics of some of them.
Intervention program rule for victims: When the participant is an offender, he
has access to any program described in the basic ontology. However, when the
participant is a victim, potential routes are: citizenship, psychosocial and psychiatric;
Rule to retrieve evidence-based solutions centered on the judge's specialty:
from the expertise of a judge, logged in the system, the research solution space can be
reduced based on: (i) his/her expertise, (ii) two specialties, or (iii) all experts;
Rule to find documents with evidence based in query keyword: the query terms
must be confronted with the words found in researched documents; programs must order
the most similar to facilitate the choice of the decision maker and a ranking as
presented.
5. Application to the Crime Prevention Scenario
We present an example adapted from a real case involving an alternative penalty - a
model for infractions that are of minor and moderately offensive potential (e.g.,
contravention, illegal weapon possession). It deals with a new modality, face-to-face
restorative justice, in which a victim that suffered violence of an alcoholic offender
receives support. A prototype, developed in Java language, interacts with a XML
Database generated by Protégé ontology editor. The original data, approximately one
20
hundred cases, were extracted from conventional Court’s Database. Figure 4 presents
data for searching by evidence in the local database.
The High or Moderate Intervention Complexity is due to offender and victim
need of treatment. The Judge’s expertise in the new case is “drug crimes”. We applied
similarity cosine formula used in Information Retrieval for keyword similarity search
between query and document with evidence.
Figure 4. Data for searching evidence from local database
In the first retrieval, we do not use contextual elements and the results with
several cases are present in Figure 5a. Using contextual information parameters as filter
fewer cases were selected (see Figure 5b). This filtering was carried out as follows: (i)
based on the desired expertise ("drug crimes" and "crimes against women") only the
documents 1, 3, 4 and 5 were selected initially; (ii) document 4 was rejected by the
safety indicator = 60.0 (so less than 70 % desired); and (iii) document 5 was not
accepted by expectation indicator = 70.0 (so less than 80 % desired).
Figure 5. Retrieved documents with evidence: a) without using context (upper),
(b) using contextual element (lower)
The presented cases are not sufficient to give support to the solution (they do not
treat face-a-face meeting). So, the judge should search for documents with evidences.
The research began with the question containing the problem and actor (woman with a
psychological problem who was assaulted), intervention (face-to-face sessions),
comparison of interventions (face-to-face sessions and conventional processes) and
outcome (beneficial effects). The sources Campbell Collaboration and Springer Verlag
were chosen and their respective home-pages were obtained. Figure 6 show data for
second search regarding documents published between 2005 and 2010.
21
Figure 6. Data for searching evidence in Springer Verlag’s database
Figure 7. a) Evaluate the best evidence; e b) Decision-making g evidence in Springer
Verlag’s database
Figure 7a shows a meta-analysis study (taken from Springer Verlag) that was
selected by the judge as presenting the best evidence on face-to-face meetings between
victims and offenders. The study sample is derived from two randomized controlled
trials: one conducted with offenders who committed crimes against private property
involving violence in Canberra, Australia, and the other, crimes of burglary with victims
in London, England. The sample context was analyzed and contrasted with the new
problem. The evidence drawn from this showed objectively that 76% of victims were
satisfied with the results obtained from the face-to-face meeting with offenders. This
study led to a successful implementation of a training course for police officers, in
which the concepts of restorative justice and practice sessions in face-to-face meeting
between offenders and victims were applied. The document and the evidence were
evaluated in terms of validity, relevance and applicability, and the information was
extracted manually and recorded in a local database.
The decision making is presented in Figure 7b. Data of the victim were informed
and they are compatible with the best evidence founded. The victim agrees to participate
22
in face-to-face meetings with the offender, provided that in previously established time
and with the presence of authorities. Victim support programs, with respect to
psychosocial and psychiatric treatment, must be offered in this particular intervention.
The process concludes with documentation of the research performance made by judge.
This example shows that the presented application has potential to be leveraged
to support a more appropriate evaluation of the ontology.
6. Related Works
In this section we present some related work on the themes evidence, context, ontology,
and integration of this themes.
In Stolba et al. (2009) is showed how Data Warehouse facilitating Evidence-
Based Medicine can be applied for reliable and secure processing of huge amounts of
medical data. The authors present a data model for building a federated Data Warehouse
considering adopted international standards for the exchange of healthcare data. Nakaya
e Shimuzu (2006) present the Knowledge representation architecture based on Evidence
based Logical Atomism (KELA) that consider the anatomic hierarchic structure from
genome to human. Knowledge atoms of molecular and disease findings are modeled as
entities and relationships - describes species, birthplace, and existing place in an
ontological view.
Vieira et al. (2010) presents a domain-independent context meta-model, which
guides context modelling in different applications. The meta-model offers integrated
support for modeling structural and behavioral aspects involved in context management
and usage. Contextual graph and UML were used. Sheng and Benatallah (2005)
Introduce the ContextUML meta-model developed to support the modeling of context-
aware Web Services. It separates modeling context (types, sources, etc.) from modeling
context-awareness (objects and Mechanisms) becoming restrict to the Web Services
category of Context-Sensitive Systems.
The related works above regard individually evidence or context. The
combination of research evidence with context was not developed computationally.
Besides, none of them has the perspectives of integration and extension for several
domains, and none of these present a vision of combining ontological proposal.
7. Conclusions and Future Work
This article proposes the integration of context with evidence represented in a meta-
model to facilitate the development of applications centered in EBP considering context
for several domains. The class structure of the meta-model was the base for build
domain ontology oriented to crime prevention. Contextual information related to the
EBP of the criminal area were represented and instantiated. With a practical
implementation we showed how contextual EBP can be used to support Judge’s
decision making and was verified that using contextual information makes the retrieve
more effective.
Future researches encompass (i) the building of: task ontology for the criminal
area; a high-level ontology for the areas that use EBP such as Medicine and Education;
and a semi-automatic Evidence-Oriented Information Extractor (EOIE); and (ii) the
23
incorporation of the classical case structure (problem, solution and result) and Case-
Based Reasoning technique for decision making support.
References
Brézillon, P. (2007). Context modeling: Task model and practice model. CONTEXT-07,
LNAI 4635. pp. 122-135, Roskilde, Denmark.
Bunningen, A. (2004). “Context Aware Querying - Challenges for data management in
ambient intelligence”. Doctorate thesis, University of Twente.
Chomsky, N. (1965). Aspects of the Theory of Syntax. MIT Press.
Dey, A.K. and Abowd, G.D. (2001). A Conceptual Framework and a Toolkit for
Supporting the Rapid Prototyping of Context-Aware Applications. Human-Computer
Interaction (HCI) Journal, v. 16, n. 2-4, pp. 97-166.
Dobrow, M.J., Goel, V. and Upshur, R.E.G. (2004). Evidence-based health policy:
context and utilization. Social Science & Medicine, Jan, 58(1), 207-17.
Friedland, D. J., Go, A.S., Davoren, J.B., Shlipak, M.G., Bent, S.W., Subak, L.L. and
Mendelson, T. (1998). Evidence-Based Medicine: A Framework for Clinical
Practice. NY:McGraw-Hill.
Gomes, G.L.R. (2008). A Substituição da Prisão – Alternativas penais: legitimidade e
adequação. Salvador: Editora Podium.
Nakaya, J. and Shimizu, T. (2006). Knowledge Architecture based on Evidence Based
Logical Atomism for Translational Research. International Journal of Computer
Science and Network Security – IJCSNS, February, v. 6, n. 2A, pp. 175-179.
Sackett, D.L., Straus, S.E., Richardson, W.S., Rosenberg, W. and Haynes, R. B. (2001).
Evidence-based medicine: how to practice and teach EBM. Elsevier Health Sciences.
Saliba, M.G. (2009). Justiça Restaurativa e Paradigma Punitivo. Curitiba: Juruá Editora.
Satterfield, J.M., Spring, B., Brownson, R.C., Mullen, E.J., Newhouse, R.P., Walker,
B.B. and Whitlock, E.P. (2009). Toward a transdisciplinary model of evidence-based
practice. The Milbank quarterly. Blackwell Publishing, June, v. 87, n. 2, pp. 368-390.
Sheng, Q. Z. and Benatallah, B. (2005). “ContextUML: A UML-Based Modeling
Language for Model-Driven Development of Context-Aware Web Services”. In Proc.
of the International Conference on Mobile Business (ICMB05), pp. 206-212.
Stolba, N., Nguyen, T.M. and Tjoa, A. (2009) “Data Warehouse Faciliting Evidence-
Based Medicine”. In Nguyen, T.M. (Ed.). Complex DW and Knowledge Discovery
for Advanced Retrieval Development, pp.174-195. Premier References Source.
Thomas, G. and Pring, R. (2004). Evidence-Based Practice in Education. Open
University Press.
Vieira, V., Tedesco, P. and Salgado, A.C. (2010). Designing Context-Sensitive Systems:
An Integrated Approach. Expert Systems with Applications, 38:2. pp.1119-1138.
Wang, X.H., Zhang, D.Q., Gu, T. and Pung, H.K. (2004). “Ontology based context
modeling and reasoning using OWL”. Proceedings of the Second IEEE Annual
Conference on Pervasive Computing and Communications Workshops, 2004.
Warren, R.K. (2007). “Evidence-Based Practice to Reduce Recidivism: Implications for
State Judiciaries”. http://works.bepress.com/roger_warren/1. April, 2009.
24