International Journal of Information System and Engineering

Vol. 6 (No.2), November, 2018

ISSN: 2289-7615
DOI: 10.24924/ijise/2018.11/v6.iss2/37.44 www.ftms.edu.my/journals/index.php/journals/ijise

This work is licensed under a

Creative Commons Attribution 4.0 International License.

Research Paper

KNOWLEDGE BASED ARCHITECTURE FOR MENTAL HEALTH CARE

SYSTEM

Nur Farah Hannan Ahmad1

Department of Information Systems
Faculty of Computer Science & Information
Technology
nurfarahannan@gmail.com

Liyana Shuib2
Department of Information Systems
Faculty of Computer Science & Information
Technology
liyanashuib@um.edu.my

Mohd Khalit Othman3

Department of Information Systems
Faculty of Computer Science & Information
Technology
mkhalit@um.edu.my

Abstract

It is important for experts to understand patient emotional state and personality traits to suggest
intervention materials for them. Different patient might have same emotional state, but different
personality; hence a tool that can help match intervention materials with the user emotional state
and personality traits is needed. Intervention materials that best suit for user’s personality is
important as it will encourage them to follow the recommended materials. Personalization can be
identified using the emotional states and personality traits. In mental health care, identifying
emotional state, and matching intervention material based on personality traits, is critical for
patient as it may affect their mental health condition progress and their willingness to follow the
recommended materials. Therefore, it is crucial for users to be able to locate intervention material
that best matches their personality traits. The objective of this paper is to develop a tool that can
help retrieve intervention material based on personalization. By using a knowledge-based method,
this tool will be able to help user match intervention materials that best suits their emotional state
and personality. The architecture and components of the tool are discussed.

Key Terms: Knowledge based, Architecture, Mental Health Care, Personality

1. Introduction

The numbers of mental health patients are increasing as years passed. Mental disorder
such as depression, anxiety and stress are an increasing problem in our society. 2018 theme
for World Mental Health Day is “Young people and mental health in a changing world”.
According to WHO, half of all mental illness begins by the age of 14, but most of the cases go

ISSN: 2289-7615
Page 37
 undetected and untreated. Meanwhile, in Malaysia, National Health and morbidity Survey
2015 stated that every 3 from 10 adults aged 16 and above have mental illness. This finding
shows mental illness not only affects adults but it also has impact to youngster as early as
14. To solve the problems and improve the quality of mental health care in Malaysia, MOH
focus on elements, which are fundamentally crucial for quality healthcare, which are
appropriate, accessible, safe, responsive, patient-centered, effective and affordable.

However, most intervention materials for mental illness only focus on the patient‘s
emotional state. The intervention material provided might not suited certain patients hence,
may cause the slower patient’s progress as a result. Habibi et al. (2013) findings show that
there is a significant and positive relationship between personality traits and dimensions of
mental health tests. Thus, this study aims is to introduced a system that not only take the
patients emotional state into account but also focus on the patients personality traits when
providing intervention materials for the patients.

Currently, there are no tools that can assist patient to get suitable intervention material
based on the patient’s emotional state and personality traits. To assist the patients finding
suitable intervention material, a knowledge-based recommender system is developed.
Personalization can be identified using both mental test and personality test.

2. Literature Review

Intervention materials for mental illness are normally given to the user without taking
into account the user’s personality traits. Examples of the intervention materials are, yoga,
hiking and games. Each personality may have different preference and liking. Each material
provided may affect the user willingness or desire to follow the recommended materials.
Suitable intervention materials recommendation plays important role; because it increase
the user willingness to follow the materials suggested and effectively improve the user
emotional state.

With increasing access to the Internet, the amount of information is growing

exponentially; thus leading to information overload problems. For this reason, users are
faced with difficulties of obtaining suitable intervention material. This is largely because
users are unaware of their own personality traits. To eliminate this problem, intervention
materials are matched depending the users emotional state and personality traits.

To evaluate the user’s emotional state, the Depression Anxiety Stress Scale (DASS) is
used. However, since the system will be a web-based system, the DASS 21-item version will
be adopted. Based on the previous study, DASS 21-item also have similar consistency,
excellent convergent validity and also good discriminative validity. Overall DASS 21-item
works as well as other commonly used measures (Gloster et al., 2008). DASS 21-item
assessment is used in the system to evaluate the user condition which are, depression,
anxiety or stress. The level of each user’s condition also can be determined.

Personalization can be applied via personality traits. In this study, the TIPI (Ten Item
Personality Measure) by Muck et al., (2007) is used, because it uses five-factor model (FFM)
or also known as Big Five. Big Five is one of the predominant models of personality in trait
psychology to date. Basically TIPI is the short version of Big Five, which only have 10
questions in the assessment. The benefits of using this assessment are the short time to
complete the questionnaire, cost effective and verified good as the Big Five measure (Muck
et al., 2007). The traits evaluate in the assessment are extraversion, emotional state,
openness, agreeableness and conscientiousness.

More alternative should be developed or introduced to cater to these people especially

youngster and young adults. To reach out to more people, mobile mental health treatment

ISSN: 2289-7615
Page 38
 applications are introduced. These applications are regarded as a promising solution to
meet increasing demands in mental health care division.

1) Mental Health Care

Another solution to cater for the demand for mental health treatment can be
online treatment (Becker, 2016; White et al., 2001). Computer-based cognitive behavior
therapy programs have proven clinically effective to treat variety of mental disorder
(Becker, 2016;Cuijpers et al., 2009; Andersson & Cuijpers, 2009; Olatunji et al., 2010).
Another advantage of using online treatment is its time and cost effective. In this
personalize system, intervention materials are selected based on the user’s emotional
state and personality traits. Each material is matched to the results of both DASS 21 and
TIPI assessments.

2) Recommender system

One of the recommendation systems purpose is to deal with information

overload (Konstan & Riedl, 2012). Information system filtering works by generating the
result depending on the user preferences and interest or by monitoring the user action
towards the recommended results or items (Isinkaye et al., 2015). According to
Isinkaye et al. (2015), collaborative filtering technique was the most mature and the
most commonly implemented. However, for this system we use knowledge based
filtering (KB). KB filtering use knowledge about users and products to pursue a
knowledge-based approach to generate recommendations and reasoning about what
products meets the user’s requirements (Bobadilla et al., 2013; Burke 2000). KB also
works best to solve the cold start problem (Bobadilla et al., 2013) which a problem that
faced by a lot of recommender system.

3. Methodology

The methodology of this research study consists of four phases (as shown in Figure 1):

Figure 1

1) Identifying Mental Test

DASS 21-item (Gloster et al., 2008) is used because it is verified to accurately

evaluate the user’s emotional state. It is also have the short version, which made it
easier for the user to answer without having to rush to complete the assessment. DASS
21 able to evaluate the user depression, anxiety and stress level. The levels for each
category are normal, mild, moderate, severe and extremely severe.

2) Identifying Personality Traits Model

The TIPI (Muck et al., 2007) was used because of it is the short version from the
Big Five model which both are verified to be able to determine the user’s personality.
The TIPI consists of extraversion, emotional state, openness, agreeableness and
conscientiousness.

3) Mapping Intervention material onto the Emotional State and Personality Traits
ISSN: 2289-7615
Page 39
 The materials provided by the experts and mapped by the experts. Extensive
research was done to map the material accordingly. First, the materials were map to the
right emotional state condition, and then only the materials were mapped to the
personality traits.

4) Designing and Developing the Tool

The system was designed and developed using a web base system. Figure 2
shows the architecture of the system. The architecture has five components, which are
Input, Output, Database and Knowledge based. Each component is discussed in the next
section.

Figure 2

5) Evaluating the Tool

The evaluation will be using Technology Acceptance Model (TAM) and Precision
and Recall.

4. 2Heal

The system named 2Heal. The modules available in Figure 2 are discus below:

1) Input Module

The module requires the user to provide their basic information such as their
name and email. Then the user will need to take the assessments, which are DASS 21
and TIPI. All the user information is stored in their user profile. Information from the
assessments is needed to provide best results for them.

ISSN: 2289-7615
Page 40
 Figure 3: User registration

Figure 4: User login

Figure 5: DASS 21-item assessment

ISSN: 2289-7615
Page 41
 Figure 6: TIPI assessment

2) Process Module

This module map and match the intervention material for the user based on
their results from both DASS 21 and TIPI assessments. The system uses the knowledge
from the user (assessments results) and generates the intervention materials
accordingly. To get the user’s feedback, this system used the user rating and user
behavior technique. The user behavior was based on the items clicked by the user.

3) Database Module

There are two database used in this system. The first database is the user profile,
which contains the user’s information and assessments results. Another database
contain the intervention materials.

4) Output Module

Assessments results and intervention materials that matched the user’s

assessment results. The user also can take the assessments more than one time to see
their mental health progress.

Figure 7: User results

ISSN: 2289-7615
Page 42
 Figure 8: User progress update in graphical format

5. Conclusion

In order to provide intervention materials for user who may or may not have mental
disorder, the user personality traits should be taken in consideration as well as the material
provided may not suitable for certain user. The implication of the wrong material being
used to people with mental disorder may affect their mental health progress and may give
negative impact in their daily life. In the future, the knowledge-based system can be
integrated to enhance the system capability and improve the recommender results.

Grant

This work was supported by the University Malaya Research Grant Programme - SBS
(Equitable Society) [RP057C-17SBS].

Reference

1 Andersson, G., Cuijpers, P., 2009. Internet-Based and Other Computerized Psychological
Treatments for Adult Depression: A Meta-Analysis. Cognitive Behaviour Therapy 38, 196–205.
doi:10.1080/16506070903318960.

2 Becker, D., 2016. Acceptance of Mobile Mental Health Treatment Applications. Procedia
Computer Science 98, 220–227. doi:10.1016/j.procs.2016.09.036.

3 Bobadilla, J., Ortega, F., Hernando, A., Gutiérrez, A., 2013. Recommender systems survey.
Knowledge-Based Systems 46, 109–132. doi:10.1016/j.knosys.2013.03.012.

4 Burke, R., 2000. A Case-Based Reasoning Approach to Collaborative Filtering. Lecture

Notes in Computer Science Advances in Case-Based Reasoning 370–379. doi:10.1007/3-540-
44527-7_32.

5 Cuijpers, P., Marks, I.M., Straten, A.V., Cavanagh, K., Gega, L., Andersson, G., 2009.
Computer-Aided Psychotherapy for Anxiety Disorders: A Meta-Analytic Review. Cognitive
Behaviour Therapy 38, 66–82. doi:10.1080/16506070802694776.

6 Gloster, A.T., Rhoades, H.M., Novy, D., Klotsche, J., Senior, A., Kunik, M., Wilson, N.,
Stanley, M.A., 2008. Psychometric properties of the Depression Anxiety and Stress Scale-21 in
older primary care patients. Journal of Affective Disorders 110, 248–259.
doi:10.1016/j.jad.2008.01.023.

ISSN: 2289-7615
Page 43
7 Habibi, Z., Sadeghi, H., Haghrangbar, F., Madanipour, K., Azarnoosh, A., 2013. The Study
of Personality Characteristics and Mental Health in Addicts. Procedia - Social and Behavioral
Sciences 84, 509–513. doi:10.1016/j.sbspro.2013.06.594.

8 Institute for Public Health (IPH) 2015. National Health and Morbidity Survey 2015
(NHMS 2015). Vol. II: Non-Communicable Diseases, Risk Factors & Other Health Problems;
2015.

9 Isinkaye, F., Folajimi, Y., Ojokoh, B., 2015. Recommendation systems: Principles,
methods and evaluation. Egyptian Informatics Journal 16, 261–273.
doi:10.1016/j.eij.2015.06.005.

10 Konstan, J.A., Riedl, J., 2012. Recommender systems: from algorithms to user experience.
User Modeling and User-Adapted Interaction 22, 101–123. doi:10.1007/s11257-011-9112-x.

11 Malaysian Healthcare Performance Unit, Malaysian Mental Healthcare

Performance:Technical report 2016, Ministry of Health Malaysia: Putrajaya. p. 1-67.

12 Mental Health. (2018, October 10). Retrieved from

http://www.who.int/mental_health/en/.

13 Muck, P.M., Hell, B., Gosling, S.D., 2007. Construct Validation of a Short Five-Factor Model
Instrument. European Journal of Psychological Assessment 23, 166–175. doi:10.1027/1015-
5759.23.3.166.

14 Olatunji, B.O., Cisler, J.M., Deacon, B.J., 2010. Efficacy of Cognitive Behavioral Therapy for
Anxiety Disorders: A Review of Meta-Analytic Findings. Psychiatric Clinics of North America 33,
557–577. doi:10.1016/j.psc.2010.04.002.

15 White, L.A.E., Krousel-Wood, M.A., Mather, F., 2001. Technology Meets Healthcare:
Distance Learning and Telehealth. The Ochsner Journal 3, 22–29.

IJISE is a FTMS Publishing Journal

ISSN: 2289-7615
Page 44