USER PROFILE-BASED SERVICE DISCOVERY for eHEALTH

Tesfa Tegegne

, Benjamin Kanagwa

and Theo van der Weide

Institute for Computing and Information Sciences, Radboud University, Nijmegen, Netherlands

School of Computing and Information Technology, Makerere University, Kampala, Uganda

Keywords:

Personalization, eHealth, Service Discovery, SOA.

Abstract:

Personalization is not very common in Service-oriented Architecture and Service discovery in practice, yet

very important as it enables to ﬁnd more speciﬁc and relevant services. This is especially true in the case of

complex service, where the system tries to ﬁnd an optimal combinations of services that together can satisfy a

(complex) request. We describe the relation between personalization and service discovery in general, and then

concretize this in the proposed Service Discovery Framework (SDF). Personalization effectively improves the

quality of the discovery algorithm in SDF. As an application, we show its beneﬁts in the context of eHealth.

1 INTRODUCTION

Personalization is regarded as one of the most com-

pelling features of service-oriented systems (Papa-

zoglou et al., 2007) and mobile computing (see (Wang

et al., 2008)) disciplines. Personalization refers to a

set of preferences, data, rules and settings that are

used by a device or service to deliver a customized

version of capabilities to the user (Kobsa, 2007; So

et al., 2009; Wagner et al., 2002) as a means of pro-

viding user-centered services. User-centered and per-

sonalization support customers in selecting their pre-

ferred service from the rapidly increasing publishing

of services.

In the past few years, personalization has been

used in various application areas such as e-commerce,

e-learning to assists users in ﬁnding, adapting and us-

ing services that fulﬁl their needs, given their personal

proﬁle, mobility and context (see (Lankhorst et al.,

2002)). Personalization is concerned with matching

and negotiation between user requirements and abili-

ties on the one hand and service offerings and result-

ing adaptation of network and application level ser-

vices on the other hand (Lankhorst et al., 2002).

In the realm of service discovery, we see person-

alization as an advanced form of service discovery: a

set of services that meet the core requirements of a

user are further scrutinized for compliance with the

user personal desires and constraints. Consequently,

personalization offers a quality user proﬁle and an ac-

curate representation of the user’s physical and emo-

tional behavior (Lankhorst et al., 2002). To build and

maintain effective personalization, appropriate user

preferences, interests, and behaviors are required (see

(Ghosh and Dekhil, 2009)). In a way, personalization

covers all functionality regarding discovery and man-

agement of users with respect to their preferences, de-

mands and wishes. Therefore, it is important that per-

sonalization should be transparently incorporated as

one of the main components of service discovery in

addition to context-awareness and semantics.

A user proﬁle is often used to classify users into

predeﬁned user segments (such as, by demograph-

ics or tastes) or to capture their behavior including

their private interests and preferences (see (Shmueli-

Scheuer et al., 2010)). One of the key challenges

pointed by (Ramanathan and Kapoor, 2009) is to con-

struct accurate user models containing demographic

interests, preferences, intent and behavior informa-

tion. A user proﬁle can be obtained in various ways.

It can be explicitly deﬁned by the user, such as dur-

ing registration for a service. Additional information

about the user can be obtained through user proﬁling:

the process of implicitly learning user characteristics

from data associated with that user. Data sources for

user proﬁling include among others the user browsing

sessions, the user’s own generated content, the user’s

social interactions with friends in the user’s social net-

work (e.g. the user’s discussions with others), click-

through data extracted from search logs, or even other

user proﬁles using collaborative ﬁltering techniques

(Shmueli-Scheuer et al., 2010). Personalized systems

employ various personalization methods like machine

learning, information retrieval, bayesian networks see

(Wang, 2010; Kobsa, 2007) to derive additional as-

sumptions about users. Generally, the more informa-

337

Tegegne T., Kanagwa B. and van der Weide T..

USER PROFILE-BASED SERVICE DISCOVERY for eHEALTH.

DOI: 10.5220/0003630003370346

In Proceedings of the 6th International Conference on Software and Database Technologies (ACT4SOC-2011), pages 337-346

ISBN: 978-989-8425-76-8

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

tion is collected about users, the better the quality of

personalization will be. As a result, the process of

personalization usually consists of user modeling and

adaptation. User modeling (Wang, 2010) is concerned

with building user models, while adaptation creates

personalized services based on the user proﬁle.

The contribution of this paper is formalized

context-aware proﬁle-based service discovery with

particular emphasis to eHealth services. In addi-

tion, we address user authentication and authoriza-

tion methods for user proﬁle and services. We as-

sess the impact of personalization on facilitating ser-

vice discovery, service selection and service match-

ing. We exemplify our work within the domain of

eHealth services motivated by the fact that it is not

common to ﬁnd identical health cases with similar

context, particularly in developing countries that are

usually synonymous with wider variations in culture,

beliefs and general living conditions. Such variations

occasions added importance for personalization espe-

cially in eHealth services that are core to developing

countries.

This paper is organized as follows. Section 2 ex-

plains related work. We overview personal proﬁle in

section 3. In section 4 we present the proposed service

discovery framework. We discuss personalization on

eHealth in section 5. Finally, we conclude in section

6 where we highlight some of the future research di-

rections.

2 RELATED WORK

the last couple of years various researchers have

addressed the problem of personalization for dif-

ferent applications (for example e-learning and e-

commerce, etc). The rapid growth and deployment of

services and information makes personalization one

of the important research topics. However, the appli-

cation of personalization in service discovery has not

been widely studied. Semantic and contextualization

of service was given more emphasis in the process of

service discovery than personalized service discovery.

The concept of a user proﬁle (personalization)

usually refers to a set of preferences, information,

rules and settings that are used by a device or ser-

vice to deliver a customized version of capabilities to

the user ((So et al., 2009)). To capture user prefer-

ence and interest, ontology based user modeling has

widely been adopted. Ontology helps to develop an

effective mechanism for user query contextualization

based on both current and previous search history (So

et al., 2009).

According to Petersen (Petersen et al., 2009), a

user proﬁle contains information on the goals, the

needs, the preferences and the intentions of that user.

The user proﬁle provides an interface to access this

user information, including its static proﬁle, prefer-

ences, context information (location, presence, device

capabilities, etc), service speciﬁc personal informa-

tion and the history of service usage and so on. The

provided function will be used to discover the services

requested by the user.

Typically a user proﬁle includes her/his individ-

ual preferences together with technical constraints of

his/her means of communication (devices and equip-

ments) and his environment (Wagner et al., 2002).

Based on semantically enriched service descriptions

a service request management is then performed in-

cluding a service discovery and execution. During

service discovery and selection, decisions that have

to be made will not necessarily lead to one deﬁnitive

outcome. Requirements could be met not by just one

service or service component, but by a set of them, or

none at all (Wagner et al., 2002). Therefore (Wagner

et al., 2002) state that the process of service discovery

can take the personal user preferences, semantically

rich user proﬁle and context information of the user

into account.

2.1 Context-aware Models for

Personalization

Context-awareness refers to systems that are able to

adapt their services to the user context. Services can

be selected according to both the current user con-

text and her/his preferences for this context. Context-

aware proﬁles aim at indicating which service is rel-

evant for a user when he is in a given situation. The

main idea of the context based proﬁle (Mehta et al.,

2005) is to allow users to deﬁne the proﬁles and po-

tential context in which these proﬁles should be used.

Context can be thought of as the “extra” often im-

plicit, information (i.e. associations, facts, assump-

tions), which makes it possible to fully understand an

interaction, communication or knowledge representa-

tion (Mehta et al., 2005). They propose the use of a

common ontology based user context model as a ba-

sis for the exchange of user proﬁles between multiple

systems and, thus, as a foundation for cross-system

personalization.

In our model the user context can be obtained in

two ways: the user announces her current proﬁle in

her request or the context manager dynamically gen-

erates the users context (location, weather, etc) from

the context provider devices such as GPS, sensors etc.

A user context (as summarized in Figure 1) con-

sists of two categories (static and dynamic). The static

ICSOFT 2011 - 6th International Conference on Software and Data Technologies

338

category describes the demographic data of the mo-

bile user. These are user data about name, gender,

age, language etc. The dynamic further classiﬁed as

environmental and information world. The environ-

ment category provides situational context informa-

tion like the actual location and the environment. The

third category captures the user’s “information world”

e.g. read documents and visited Web pages - thus re-

ﬂecting the user’s interests (Kuck and Gnasa, 2007).

Figure 1: Context of mobile user (Kuck and Gnasa, 2007).

Some of the advantages of using context-

awareness in healthcare domain are large number

of situations and related tasks, mobility of patients,

health professionals and some equipments; limited ﬁ-

nancial and human resources; and the need to cut cost

while improving the quality of service to an increase

number of people. In addition, the expectation to ac-

cess, process, and modify the healthcare information

anywhere and anytime using handheld devices is an-

other reason to use context-awareness. Alike other

systems, contextualization encounters different chal-

lenges such as how often the contextual information

needs to be collected, when is some information is

old, not reliable, missing, or just plain wrong? Is it

the same for normal, abnormal and emergency cases?

Health professionals and patients are involved in a

variety of tasks and processes and their information

needs vary based on their location, time, environment

and involved activity. The knowledge of their current

activity may assist in deciding which of the health-

care professionals should be reached for some medi-

cal tasks or decision making.

2.2 Personalization and Service

Discovery

Personalized search is seen by the research commu-

nity as a means to decrease search ambiguity and re-

turn results that are more likely to be interesting to

a particular user and thus providing more effective

and efﬁcient information access (Sieg et al., 2007;

Park et al., 2009). One of the key factors for accu-

rate personalization information access is user pro-

ﬁle and user context. There are many factors that

contribute for the delineation of personalization such

as: the user’s short-term information need, such as a

query or localized context of current activity, seman-

tic knowledge about the domain being investigated,

and the user’s proﬁle that captures long-terminterests.

To ﬁnd the most appropriate services, service dis-

covery protocols should exploit context. The most

appropriate service is not always the nearest or the

highest quality rather, it varies according to the user’s

preferences. More importantly, user criteria change

with their context. A user might generally prefer to

use the most popular services, but in a particular case,

might want to use only high-quality services (Park

et al., 2009). A service discovery starts from the local

service repository/registory. If it fails to ﬁnd an ap-

propriate service in the local repository or registry, it

propagates request/discovery queries to other appro-

priate repositories/registries. If the requested service

is found, the matchmaking process will be performed

to select the appropriateservice based on the user con-

text and user preference (proﬁle) (Park et al., 2009).

Discovery corresponds to the activity of locating

a service that meets certain functional criteria. Selec-

tion is related to the activity of evaluating and ranking

the discovered services to identify the ones that fulﬁll

a set of non-functionalproperties and user preferences

requested by the actual user. Most of the existing

techniques rely on syntactic descriptions of services

interfaces to ﬁnd services with disregard to non func-

tional service parameters (Badr et al., 2008). Incorpo-

rating user contexts and user preferences in the pro-

cess of service discovery brings quality service dis-

covery that can improve search results and provides

a better means for user to identify their service infor-

mation needs (Badr et al., 2008; Park et al., 2009).

3 USER PROFILE

The main purpose of user proﬁle is to keep track of

the user related information to control its availability

and facilitate service access for its users, and to real-

ize a better user experience. Besides, personalization

helps to identify a service delivery mechanism (text,

audio, video, etc.), to choose the type of service to

be delivered with respect to user preferences (physi-

cal, psychological, environmental, and so on) and to

create a global personalization scheme (ETSI, 2005).

User proﬁle may contain many individual data items

(information, preference and rules) coming from dif-

ferent sources (Choi and Han, 2008; Petersen et al.,

2009; ETSI, 2005). When a user wishes to have the

behavior of services personalized to his/her require-

ments a proﬁle will be required (ETSI, 2005). The

USER PROFILE-BASED SERVICE DISCOVERY for eHEALTH

339

value of some of these items will (1) either not change

(static) or (2) change very irregularly and (dynamic)

and (3) some may be individuallychanged by the user.

A single user may be associated with multiple pro-

ﬁles. It is therefore important that users have the ﬂex-

ibility to decide on the proﬁle to use or better, the

system should automatically detect the current pro-

ﬁle. For instance, a user may move from ofﬁce to

a meeting room requiring him to access services on

a small device. Users can often control the type of

service provided, as well as the look and feel of the

interface, by indicating their choice through their pro-

ﬁle. In (Petersen et al., 2009) a user plays a number

of roles in personalization system. A typical user in

personalization will have at least two roles such as the

user role and the administrator role.

User proﬁle can be segmented into private and

group (clustered). The private proﬁle contains only

individual data i.e. personal data, and preferences.

Group proﬁle is generated from the private proﬁle

from the users who share similar interests and pref-

erences. For example, for healthcare works we can

group users based on their job or educational sta-

tus: specialist, general practitioners, nurses, pharma-

cist, etc.; and patients can be grouped or clustered by

the disease type: HIV patient, TB patient, diabetic,

etc. This type of clustering improves the service se-

lection and service matchmaking process since they

share similar information. Furthermore, this segmen-

tation assists to push (broadcast) some information to

the users with the same preferences (for instance, for

healthcare workers new disease break out, new vac-

cination, training etc. and for patents medical time,

consultation and education etc).

User proﬁle determines the method of service de-

livery. For example a vision impaired person wishes

to get the service in audio and the hearing impaired

user wishes to receive in text rather than audio (Choi

and Han, 2008; Petersen et al., 2009; ETSI, 2005).

Additionally, user proﬁles may require to be priori-

tized to avoid conﬂicting proﬁles. A user may have

several preferences, however, services might not be

available for the preferences to match. Hence, after

ﬁltering out the relevant needs of the users prefer-

ence required to be prioritized. The priority of user

preferences will help to relax the matchmaking pro-

cess. In line to this (Petersen et al., 2009; ETSI, 2005)

stated eHealth proﬁles to be allocated higher priorities

than non-eHealth proﬁles. eHealth proﬁles are con-

sidered as more important than personal proﬁles (nor-

mal proﬁles) as they contain information (extracted

from electronic health records) and preferences (Choi

and Han, 2008; Petersen et al., 2009; ETSI, 2005).

Generally, the user proﬁle should address the fol-

lowing questions: where a user is located, what her

job/ profession is, what her preferences (interest) is,

what the context of the user is, who the provider of

the service is, the cost and availability of the service

(Petersen et al., 2009; ETSI, 2005).

3.1 Challenges of User Proﬁle

Personalization (personal proﬁle) is not without chal-

lenges. Some of the challenges are: how can per-

sonal informationbe gathered, howis this information

stored in a persistent and ubiquitous manner (this is

applicable for infrastructure-less networks), how can

the changes always be reﬂected in the services ren-

dering process? In addition, privacy is the other bot-

tleneck in personalization. In regard to gathering in-

formation, we propose pull and push model as well as

manual and automatic information gathering. Man-

ually, the user will ﬁll in the form and continuously

update the proﬁle whenever there is a change whereas

in automatic information gathering especially, in mo-

bile ad hoc networks, every node can broadcast its

proﬁle to the immediate neighbor node and gathering

information from proﬁle providers and devices such

as sensors, GPS, etc. In the case of user proﬁle pri-

vacy, in our framework, we proposed user authentica-

tion and authorization model to secure the user proﬁle

from privacy threat.

3.2 User Proﬁle Privacy

Personalization is beneﬁcial for service consumers

and service providers. However, its beneﬁt may be

counteracted by privacy constraints. Personalized

systems need to address the privacy constraints, in-

cluding user personal privacy preferences, the pri-

vacy law and the regulation of the company or in-

dustry (Wang, 2010; Kobsa, 2007). Similarly, it has

been tacitly acknowledged for many years that per-

sonalized interaction and user modeling have signif-

icant privacy implications, due to the fact that per-

sonal information about users needs to be collected

to perform personalization (Kobsa, 2007). Mobile de-

vices can provide useful personalized services that are

based on usage patterns and the current user location.

Both are tracked in central servers, which creates new

privacy problems.

According to literature (Kobsa, 2007; Wang and

Kobsa, 2009; Wang, 2010), a signiﬁcant number of

people revealed their concern of disclosing personal

data to the Internet. This implies that if users are not

given guarantees of keeping their personal informa-

tion secured, then they are not willing to disclose this

kind of information.

ICSOFT 2011 - 6th International Conference on Software and Data Technologies

340

The personalized system tends to provide users to

block the privacy from any unauthorized entities. For

example the X-ray images and other important pa-

tients data can be blocked from the service providers,

even only authorized healthcare workers can get ac-

cess to the speciﬁed medical information.

An important issue of the proposed system is

user authentication and authorization. Authentication

checks the identity of the user. Authorization deals

with role based access control. Privilege will be given

to the users (human or systems) with the respect to

the role they played. For example, PERMIS (for more

details (Chadwick et al., 2008)) is one of eminent role

based access control (authorization) system used in

many applications.

3.3 Working Example

John works in non governmental charity organization.

After a few days in a charity work he feels a little dis-

comfort. He has fever, headache and nausea. The

village health extension worker (HEW) diagnoses a

laboratory examination (blood, stool, etc), however,

all the results are negative. The HEW makes addi-

tional examination to ﬁnd the cause to administer the

appropriate treatment. But the case of John cannot be

resolved in the health post as there is only one small

laboratory and there is no specialist. Thus the HEW

decides to access the health system to get additional

help. After logging into the system, the HEW ﬁlls

the form, entering the diagnosis made, the admin-

istered medication, the patient’s symptoms, the pa-

tient’s history, John’s preferences and the location of

the hospital. After accessing John’s medical history

(Electronic Health Record) remotely, the HEW learns

that John has a mild hypertension case and is aller-

gic for antibiotics, and John prefers not to have an

injection. Based on this information, some services

are retrieved from the repository. After the system re-

ceives the information, it forwards all this information

to John’s family doctor (if any) and the emergency

dispatch center (emergency dispatch center is a spe-

cialized or referral hospital near by the given location

area). This dispatch centers which has good medical

facilities and a number of specialists. While receiving

this emergency request the system will forward the

cases to the appropriate specialist to assist the patient

remotely. These emergency team will communicate

the patient’s private doctor and forward the prescrip-

tion to the attending healthcare worker to administer

the treatment.

As shown in Figure 2 a user may use two ways

of requesting a service: (1) using voice and (2) us-

ing text or/and video (depending on the bandwidth in

Figure 2: Personalized user service request.

developing countries context). As most of the popu-

lation in developing countries is illiterate, especially

in the rural areas, voice based communication will

increase the number of eHealth system users. Fur-

thermore, users who own a smart phones and can use

the features of the smart phone or PDA, can query

service using text and video (if the infrastructure al-

lows). As depicted in the Figure 2, the user can

press a button and send a voice request. The voice

server (VoiceXML) converts the voice data into text

data, and forward the converted request to the appli-

cation server. After receiving the request, the appli-

cation server will forward the request to the appropri-

ate emergency dispatch centers (EDC) based on the

location. EDC is a center which is set up based on

the geographical location. This center should have

a number of specialist doctors, good health facilities

and services. After evaluating the user request, the

EDC forwards the request to the appropriate special-

ists. The specialists evaluate the user request, verify

and access the patient’s history (EHR), communicate

with the family doctor (if any) of the patient, and ac-

cess the service repository. Finally, the treatment and

its procedure is forwarded to the mobile of the re-

questing HEW.

A voice based querying services has its own chal-

lenges. But in the case of Ethiopia it is very im-

portant since the majority of the population does not

speak English rather can they use their own language.

Choosing the language to be localized is another chal-

lenge since many languages are spoken in Ethiopia.

In our project we plan to use Amharic and English as

a medium of communication in our system. The for-

mer is understood by many of the people and is the

working language of the country and the latter is used

as a medium of instruction for secondary and tertiary

education.

4 SERVICE DISCOVERY

FRAMEWORK

In this section we ﬁrst introduce the discovery prob-

lem, and then describe the SDF framework to handle

this problem. This framework extends the description

USER PROFILE-BASED SERVICE DISCOVERY for eHEALTH

341

Figure 3: The Service discovery framework (SDF).

from previous section, see Figure 2.

4.1 The Proﬁled Discovery Problem

Service discovery is about functional matching of

atomic services. The service composition problem

can be described as the problem of ﬁnding an opti-

mal combination that together can satisfy a (complex)

request. We use the term complex discovery for the

combination of service discovery and service com-

position. We follow the approach from (Kanagwa,

2009), and shortly overview the approach. Services

are described by their effect in terms of how they

transform their input message into an output mes-

sage. We use the convenience of Hoare Logic (He and

Hoare, 1998) and write {P}S{Q} to describe that ser-

vice S will transform an input satisfying condition P

into an output satisfying condition Q. We assume a

set

of so-called elementary services. Services may

be combined to form new services. For example, the

service S

is the service in which the execution of

service S

is followed by S

. The semantics of this

compound service is derived as follows:

{P}S

{Q}⇔∃

[{P}S

{R} ∧ {R}S

{Q}]

Other ways to combine services are optional, repeti-

tive and parallel execution.

Let

contain all elementary services from

and be closed under the composition operators dis-

cussed above. A request q is described as a com-

bination q = (Pre, Post). A proﬁled request (q, P) is

the combination of a request q and a preference pro-

ﬁle P. The solution space for request q is: Sol(q) =



S ∈



{Pre}S{Post}



From this set Sol(q) the

cheapest solution is selected:

argmax

S∈Sol(q)

Pref(S, q, P)

where Pref(S, q, P) indicated how well solution S in

the context of question q is to be preferred given

proﬁle P. This problem however is computationally

intractable. In (Kanagwa, 2009) a similarity calcu-

lus is described that evaluates the similarity Sim(q, S)

between a request q and service S as an estimation

of its successfulness in satisfying request q, by in-

troducing a similarity calculus for predicates. So

Sim(Pre, Pre(S)) expresses the similarity of the re-

quested precondition Pre with the advertised precon-

dition of service S. Analogously, Sim(Post(S), Post)

expresses the similarity the advertised post condition

with the requested one. The similarity calculus is used

to select a set of candidates for further investigation to

ﬁnd a proper solution.

We will therefore deﬁne the solution space as a

mapping that indicates the correctness of solutions:

Sol(q) :

→ [0, 1] which may be deﬁned for request

q = (Pre, Post) as follows:

Sol((Pre, Post))(S)

= Sim(Pre, Pre(S)) · Sim(Post(S), Post)

Here Sim(Pre, Pre(S) expresses to what extent the pre-

condition of request q satisﬁes the precondition of ser-

vice S. In (Kanagwa, 2009) the similarity between

predicates P and Q is deﬁned as the likelihood of the

implication P⇒Q, or: Sim(P, Q) = Val(P⇒Q). For

more information on the function Val, see (Kanagwa,

2009).

Notice that in (Kanagwa, 2009) some heuristics

are described to ﬁnd a most suitable compound ser-

vice. As suggested in (Paolucci et al., 2002) we

consider three stages of capability matchmaking (in-

cluding user context, service context, user preferences

etc.):

1. Exact match: Both output and input of a service S

are equivalent to the request:

Pre(S) = Pre ∧ Post(S) = Post

2. Plug-in match: The outputs of service S are more

speciﬁc than requested, or its inputs require less

information than the requested:

Pre⇒Pre(S) ∨ Post(S)⇒Post

3. Subsumed match: The output of a service S can

only partial fulﬁll the request, or the input re-

quired by the service are more speciﬁc than the

request provides:

Pre(S) ⇒Pre ∨Post⇒Post(S)

4. Not match: if non of the above are applicable.

The ultimate goal of match making is to deter-

mine the contents that satisﬁed user preferences and

interests. For this, the matching algorithm starts with

the most speciﬁc expression which best ﬁts the user’s

wishes and in the case that no matching can be found,

the user query required to be relaxed (expanded).

ICSOFT 2011 - 6th International Conference on Software and Data Technologies

342

4.2 Preference Calculus

Input and output based matching alone can not guar-

antee the retrieved service is usable. Taking the pro-

ﬁle P into account, we get for each service S ∈

the

suitability score Suit(S, q, P):

Suit(S, q, P) = Sol(q)(S) ·Sim(P, S) (1)

The handling of user preferences may be modiﬁed

for those cases where preferences are not easily deter-

mined as exact numbers. A discrete set of values is

more appropriate for such cases. A naive approach in

such a case is to only register whether there is a pref-

erence for a service or not. However, we feel such

a boolean approach will be too restrictive for general

application. We therefore propose to use a fuzzy ap-

proach, and consider preference as a linguistic vari-

able that can take, according to the MoSCoW Prior-

itization scheme for project management (Clegg and

Barker, 2004) one of the following values: (1) Must

have (2) Should have (if at all possible), (3) Could

have (if it does not effect anything else) and (4) Won’t

have (this time but would like in the future), extended

with the value (5) Don’t want. The value of Sim(P, S)

then is expressed as such a MoSCoW value.

Referring to Formula (1), we have to combine

the outcome of Sol(q)(S) with the MoSCoW value.

A very simple approach is to convert the MoSCoW

values to numbers as follows: m(Must have) =

1, m(Should have) = 0.75, m(Could have) = 0.50,

m(Won’t have) = 0.25 and m(Don’t want) = 0. This

leads to:

Suit(S, q, P) = Sol(q)(S) · m(Sim(P, S)) (2)

4.3 The Architecture

The framework presented in our previous work

(Tegegne et al., 2010), see Figure 3, depicts an over-

all process of service discovery. A user may send a

request to the request interpreter. The proﬁle agent

personalizes the request, while the context agent will

add context speciﬁc extensions to the request. Then

the enriched request is forwarded to the matchmaker

agent. The matchmaker agent selects the best candi-

date for the submitted request as described in the pre-

vious subsection, using the service repository agent

to supply the base knowledge (their identity and their

semantics) of the elementary agents (

). The result

is transmitted to the service requester to be communi-

cated with the requester.

The proﬁle manager (PM) gets the user request.

Upon ﬁrst usage, a user may have to supply some

base information to set up a proﬁle. The proﬁle man-

ager updates the proﬁle repository. After completion,

the PM will send the user a registration conformation

message. In the meantime, the proﬁle manager sends

the new user information to the authorization and au-

thentication module. Then the user will be provided

a certiﬁcate. If the user is a registered user, the PM

will request an authorization certiﬁcate from the au-

thentication and authorization module. The PM also

checks the validity of the certiﬁcate and checks if the

certiﬁcate is given by a trust security manager. If the

certiﬁcate fulﬁlls the above requirements, the PM up-

dates the user proﬁle (preference, interest, context (if

exist)). After that the PM will send the user request

plus the current user proﬁle to the matchmaking agent

(and/or the user if necessary).

All proﬁle data are held in a proﬁle database,

which is only accessible via the PM. The PM permits

to read or write data only if the requesting application

can prove sufﬁcient access rights with a certiﬁcate.

To enhance security for sensible data, proﬁle entries

in the database are encrypted by the PM.

4.4 The Matchmaker

In this section we elaborate the match making func-

tionality from sections 4.1 and 4.2 in some more de-

tail, in terms of the architecture described in 4.3. The

function of the match maker is to ﬁnd the most suit-

able compound service S given a request q and a pro-

ﬁle P, using Suit(S, q, P) as suitability measure. The

match maker thus ensures that the user will receive

the service that is pertinent with user preferences and

interests.

Figure 4: Proﬁle-based service discovery and selection.

In Figure 4 we present the general architecture for

the match making process. First, it receives the user

proﬁle and preferences from the query interpreter and

it combines with service ontology in service reposi-

tory/registry. After that the match maker selects the

advertised (i.e. atomic) services that match with the

request. The extracted services will be forwarded to

the reasoning module to ﬁnd the best matching com-

pound services, and rank them according to their ap-

plicability. After the matching the reasoning agent

will strengthen the matching with extra criteria (QoS,

USER PROFILE-BASED SERVICE DISCOVERY for eHEALTH

343

cost etc.). Later the ranking module ranks the ex-

tracted services (matched services) based on user

preferences, user context and non functional proper-

ties of services. Finally, the services with maximal

rank will be presented to the requestor. The match

maker will further perform consistency check and va-

lidity of the user proﬁle and quality of service (Alam

et al., 2009).

Algorithm 1: ServiceDiscovery.

Request Negotiate:

GetRequest (user,req,pref,context);

Request Interpreter:

Update Proﬁle (user, req, pref out req, pref);

Update Context (context out context);

P := prefs ⊕ context;

Matchmaker:

evaluate function Sol(req);

if for all S we have Sol(req)(S) < δ then

failure;

else

return argmax

Suit(S, req,P)

end if

This discovery process is summarized in algo-

rithm ( 1) and in Figure 4. Finally some remarks

about authentication and authorization. In the archi-

tecture of Figure 4 the user requests services. The

query interpreter will send the query to the proﬁle

manger. The proﬁle manager checks whether the user

is authenticated and authorized. After that, the pro-

ﬁle manager retrieves any available user preferences

from the proﬁle repository. Finally the user informa-

tion will be forwarded to the query interpreter. The

query interpreter transfers the service request and the

user proﬁle (user data and preferences) to the match

maker to process service matching.

5 eHealth PERSONALIZATION

Advances in service orientation, mobile and ubiqui-

tous computing, and continuous progress in medical

devices and diagnosis methodology are enabling per-

sonalized healthcare services to be delivered to indi-

viduals at any place and any time (Zhang et al., 2004).

These advances deviate from the ”One size ﬁts all”

paradigm in healthcare as is common in traditional

hospitals, clinics and healthcare centers. Personal-

ized healthcare ensures that healthcare services pro-

visioned to individuals are customized to their pre-

vailing healthcare needs. In most developed coun-

tries, personalized healthcare is provided for citizens

through home care, wearable devices, online treat-

ment and consultation. In developing countries per-

sonalized healthcare could provide an opportunity to

compensate for the shortage of professional health-

care workers as well as poor healthcare infrastruc-

tures.

Introducing ICT in the healthcare sector is a hot is-

sue, both in developed and developing countries. Sev-

eral researchers have been conducting researches how

to use ICT in supporting healthcare. Many organiza-

tions are involved producing in electronic healthcare

standards. For example, Health Level 7 is a health

informatics standard organization. The main purpose

of this organization is developing, publishing and pro-

moting a comprehensive framework for the develop-

ment of health informatics standards and the employ-

ment of the framework to produce protocol speciﬁca-

tions for health data interchange, integration, storage

and retrieval among diverse data acquisition, process-

ing, and handling systems.

Similarly, the European Telecommunications

Standards Institute (ETSI) is an ICT standard organi-

zation. ETSI, recently released a ﬁrst draft personal-

ization on eHealth standard. This draft standard will

help to manage the user proﬁles for personalization

of eHealth systems and services according to the user

preferences and needs. This proﬁle is an extension of

the ‘user proﬁle management(UPM)’ standard devel-

oped by ETSI.

5.1 eHealth Service Discovery

Framework

Adapting an eHealth system to the individual user is

essential for making it safe and easy to deploy and

to use as an effective support to independent living.

Personalization can thus enhance the user’s trust in

the eHealth systems, and make them more readily ac-

cepted.

The case for personalization of health informa-

tion is supported by studies in health communication

which have shown that health-education material can

be much more effective if customized for the individ-

ual patient in accordance with their medical condi-

tion, demographic variables, personality proﬁle, and

other relevant factors (DiMarco et al., 2009). Accord-

ing to (DiMarco et al., 2009), little work is being done

on what we consider the most important factor: tailor-

ing the information to individual needs, experiences,

and communication style.

One of the most important aspects needed in sys-

tem building is information about the intended user

goals, needs, moods, preferences, intentions, etc.

This information can be acquired through an explicit

and implicit process called user modeling. Modeling

a user means interpreting user actions within a given

ICSOFT 2011 - 6th International Conference on Software and Data Technologies

344

system or application. The user model is usually re-

stricted to some characteristics that are supposed to

be the most meaningful in the context of the user in-

teraction with the system.

5.2 Key Beneﬁts

Personalization in eHealth has twofold advantages;

for the health workers it provides information (such

as training, professional guidance and assistance) on

the basis of their educational background and job de-

scriptions. On the side of the patient, it provides the

necessary patient-oriented health information systems

like patient preference, informed consent, self-care

and shared patient-doctor decision making.

The main aim of personalizing eHealth is to ac-

quire and capture factual information about the pa-

tient, their condition, current treatments, their pref-

erences. This information can be obtained from the

patient record except the user contextual information.

Personalizing eHealth besides informing, en-

abling decision making and persuading the patients,

it will give users access to medical treatment and con-

sultation services on the basis of their needs. The

health workers also beneﬁt from this personalization

in two ways, ﬁrst they strictly consult their patients

based on their preferences, and second the health

worker can fetch any information from the personal-

ized system. Patients can get information about the

treatment, background information about their condi-

tions (the cause, symptoms, its consequences and so

on) and alternative treatment and its effect.

In most of the previous personalized eHealth, the

interaction between the system and the patient was

ﬁxed and simple (Cawsey et al., 2007). Personal-

ized email or text messages have also been used.

This type of personalization does not consider indi-

vidual preferences and interests. For example if a cen-

tralized HIV/AIDS prevention and controlling ofﬁce

sends an instant message to HIV/AID patients to take

their medicine on a speciﬁed regular time. However

this system does not address the individual preference

rather it implies “one size ﬁts all” notion.

Service personalization in eHealth will help to de-

sign a personalized user interface based on user’s abil-

ities, interests and needs. This will create an oppor-

tunity in addressing heterogeneous preferences of in-

dividual users. For instance, for an illiterate mobile

user, it is possible to create an audio based user inter-

face that can be accessed by pressing a single key.

6 CONCLUSIONS

Efﬁcient and accurate service discovery is one of the

key challenges of service-oriented systems. In this

paper we concretize the use of user proﬁle provided

more reﬁned service discovery. To address the privacy

concerns that relate to use of user proﬁles, we have in-

tegrated privacy (authentication and authorization) in

service personalization. The intended deployment is

in transition countries where voice via mobile phone

is the most widespread communication mechanism.

Consequently, part of the future work is to implement

the framework using voice technology based on mo-

bile phone platform as well as developing a corpus

for local languages (such as the Ethiopian Amharic)

in the healthcare domain.

REFERENCES

Alam, S., Iqbal, Z., Noll, J., and Chowdury, M. (2009). Se-

mantic personalization framework for connected set

top box environment. In Second International Confer-

ence on Advances in Human-Oriented and Personal-

ized Mechanisms, Technologies, and Services, CEN-

TRIC, pages 97–102.

Badr, Y., Abraham, A., Biennier, F., and Grosan, C. (2008).

Enhancing web service selection by user preferences

of non-functional features. In NWESP, IEEE, 4th In-

ternational Conference on Next Generation Web Ser-

vices Practices, pages 60–65.

Cawsey, A., Grasso, F., and Paris, C. (2007). Adaptive

information for consumers of healthcare. Springer-

Verlag Berlin Heidelberg.

Chadwick, D., Zhao, G., Otenko, S., Laborde, R., Su, L.,

and Nguyen, T. (2008). Permis: A modular authoriza-

tion infrastructure. Concurrency and Computation:

Practice & Experience, 20(11):1341–1357.

Choi, O. and Han, S. (2008). Personalization of rule-based

web services. Sensors, 8(4):2424–2435.

Clegg, D. and Barker, R. (2004). Case Method Fast-Track:

A RAD Approach. Addison-Wesley.

DiMarco, C., Wiljer, D., and Hovy, E. (2009). Self-

managed access to personalized healthcare through

automated generation of tailored health educational

materials from electronic health records. In Virtual

Healthcare Interaction: Papers from the AAAI Fall

Symposium (FS-09-07,, AAAI Fall Symposium Se-

ries), pages 4–7.

ETSI (2005). Human factors (hf); user proﬁle management.

In ETSI EG 202 325 V1.1.1, ETSI Guide, pages 1–

100.

Ghosh, R. and Dekhil, M. (2009). Discovering user pro-

ﬁles. In International World Wide Web Conference,

Proceedings of the 18th international conference on

World wide web, pages 1233–1234.

He, J. and Hoare, C. A. R. (1998). Unifying theories of

programming. In RelMiCS, pages 97–99.

USER PROFILE-BASED SERVICE DISCOVERY for eHEALTH

345

Kanagwa, B. (2009). Design, Discovery and Construction

of Service-Oriented Systems. Siks dissertation series:

2009-09, Radboud University, Nijmegen.

Kobsa, A. (2007). Generic user modeling systems. In Lec-

ture Notes in Computer Scinence, Modeling technolo-

gies, The adaptive web: methods and strategies of web

personalization, pages 136–154.

Kuck, J. and Gnasa, M. (2007). Context-sensitive service

discovery meets information retrieval.

Lankhorst, M., Kranenburg, H. v., Salden, A., and Pedde-

mors, A. (2002). Enabling technology for personaliz-

ing mobile services. In HICSS-35/02, Proceedings of

the 35th Hawaii International Conference on System

Sciences, pages 87–2.

Mehta, B., Niederee, C., Stewart, A., Degemmis, M., Lops,

P., and Semeraro, G. (2005). Ontologically-Enriched

Uniﬁed User Modeling for Cross-System Personaliza-

tion. Springer Berlin / Heidelberg.

Paolucci, M., Kawamura, T., Payne, T. R., and Sycara, K.

(2002). Semantic matching of web services capabili-

ties. In ISWC 2002, pages 333 –347.

Papazoglou, M. P., Traverso, P., Dustdar, S., and Leymann,

F. (2007). Service-oriented computing: State of the

art and research challenges. Computer, 40(11):38–45.

Park, K., Yoon, U., and Ki, S. (2009). Personalized service

discovery in ubiquitous computing environments. Per-

vasive Computing, 8(1):58–65.

Petersen, F., Bartolomeo, G. andPluke, M., and Kovacikova,

T. (2009). Context sensitive personalization uti-

lizing user proﬁles: Standardization work at euro-

pean telecommunications standards institute (etsi. In

ETSI, Second International Conference on Advances

in Human-Oriented and Personalized Mechanisms,

Technologies, and Services, pages 20–25.

Ramanathan, K. and Kapoor, K. (2009). Creating user pro-

ﬁles using wikipedia. In Conceptual Modeling - ER

2009 Springer-Verlag Berlin Heidelberg, 28th Inter-

national Conference on Conceptual Modeling, Gra-

mado, Brazil, November 9-12, 2009., pages 415–427.

Shmueli-Scheuer, M., Roitman, H., Carmel, D., Mass, Y.,

and Konopnicki, D. (2010). Extracting user proﬁles

from large scale data. In ACM International Con-

ference Proceeding Serie, Proceedings of the 2010

Workshop on Massive Data Analytics on the Cloud,

page Article No. 4.

Sieg, A., Mobasher, B., and Burke, R. (2007). Web search

personalization with ontological user proﬁles. In

ACM, Proceedings of the sixteenth ACM conference

on Conference on information and knowledge man-

agement, pages 525–534.

So, C., Lai, C., and Lau, R. (2009). Ontological user proﬁl-

ing and language modeling for personalized informa-

tion services. In IEEE Computer Society, 2009 IEEE

International Conference on e-Business Engineering,

pages 559–564.

Tegegne, T., Kanagwa, B., and Weide, T. v. d. (2010). An

ehealth service discovery framework for a low infras-

tructure context. In 2nd Int. Conf. on Computer Tech-

nology and Development, Nov 2-4, Cairo, Egypt.

Wagner, M., Balke, T., Hirschfeld, R., and Kellerer,

W. (2002). A roadmap to advanced person-

alization of mobile services. In Workshops

DOA/ODBASE/CoopIS, IRvine, CA, In Proc. of the

Int. Federated Conferences DOA/ODBASE/CoopIS

(Industry Program). Irvine, CA, 2002.

Wang, H., Shi, X., Li, Y., Chang, H., Chen, W., and Tang,

J. (2008). User proﬁle management for personalized

telecom service. In IEEE Computer Society, The 9th

International Conference for Young Computer Scien-

tists, pages 1087–1092.

Wang, Y. (2010). A framework for privacy-enhanced per-

sonalization. In PhD Dissertation, University of Cal-

ifornia, Irvine, PhD Dissertation.

Wang, Y. and Kobsa, A. (2009). Performance Evaluation

of a Privacy-Enhancing Framework for Personalized

Websites. Springer Verlag.

Zhang, D., Yu, Z., and Chin, X. (2004). Context-aware in-

frastructure for personalized healthcare. The Interna-

tional Workshop on Personalized Health, December

13-15, Belfast, Northern Ireland.

ICSOFT 2011 - 6th International Conference on Software and Data Technologies

346