Predicting the misconception of dengue disease based on the awareness survey

Bulletin of Electrical Engineering and Informatics
Vol. 9, No. 4, August 2020, pp. 1631~1637
ISSN: 2302-9285, DOI: 10.11591/eei.v9i4.2390  1631
Journal homepage: http://beei.org
Predicting the misconception of dengue disease based on
the awareness survey
Abrar Noor Akramin Kamarudin1
, Zurinahni Zainol2
, Nur Faeza Abu Kassim3
1,2
School of Computer Science, Universiti Sains Malaysia, Malaysia
3
School of Biological Sciences, Universiti Sains Malaysia, Malaysia
3
Vector Control Research Unit, School of Biological Sciences, Universiti Sains Malaysia, Malaysia
Article Info ABSTRACT
Article history:
Received Nov 20, 2019
Revised Mar 29, 2020
Accepted Apr 3, 2020
Mosquito-borne disease such as dengue fever is a pervasive public health
problem around the world and further investigation is needed to rectify
the misunderstanding of the disease among communities. This requires
a personalized information delivery, which will effectively fix the problem.
The process of personalizing information requires several major steps:
(i) determine the attributes which will be used to interpret a person,
(ii) selects an algorithm which will accurately and efficiently classify the person
according to the retrieved background information, and (iii) recommends
the correct information to rectify the particular misunderstanding.
This research paper considers the first two steps. First, data regarding
the knowledge, attitudes and prevention practices are determined from
the established literature where some variables give a significant impact on
the predictive model. In the second step, five performed machine learning
algorithms were tested for the classification task. The result indicates that the use
of support vector machine and decision tree algorithms provide the best
performance in classifying the person’s understanding regarding the dengue fever.
Keywords:
Attitudes
Awareness survey
Dengue
Machine learning
Misconception prediction
Personalization
This is an open access article under the CC BY-SA license.
Corresponding Author:
Zurinahni Zainol,
School of Computer Science,
Universiti Sains Malaysia,
11800 Georgetown, Penang, Malaysia.
Email: zuri@usm.my
1. INTRODUCTION
The rate of mosquito-borne diseases (MBDs) outbreak is always increased every time the seasons
change due to various factors such as socioeconomic and environmental changes, socio-geographical,
imported cases and human movement patterns [1, 2]. All of the elements can continue the mosquito propagation
and increase the difficulties of controlling the MBDs as predicting the future consequences in a changing
environment demands a greater understanding of the array of biological and environmental features [3].
Current MBDs outbreak prediction systems were developed based on environmental factors such as flood,
types of building, daily mean temperature, daily rainfall and vegetation index [4-7]. House Index and Breteau
Index features were also often included in the predictive model to denote the risk areas [8]. Most studies have
proven that climate change gives a significant impact on the increasing risk of dengue cases by using a data
mining classification algorithm [9, 10]. Obtaining such data requires constant observation around the hotspot
area, but, the environmental monitoring system alone still cannot decrease the infection of dengue virus [1].
To improve the prediction framework, the morbidity rate forecasting works better when the vectors
like Aedes aegypti and Aedes albopictus female mosquitoes and larvae index were selected as parameters as

 ISSN:2302-9285
Bulletin of Electr Eng & Inf, Vol. 9, No. 4, August 2020 : 1631 – 1637
1632
reported in [11]. Thus, this generates an idea to include the attributes based on the human behaviors
regarding the MBDs as in [12-18] and put it into the predictive model to classify one’s misconception in
a personalized e-learning system. This research proposes the human knowledge, attitudes and preventive
practices to be included in the machine learning approach in order to identify and provide a simple MBDs
awareness program in a personalized e-learning environment. The proposed predictive framework for
the personalized e-learning environment is illustrated in Figure 1.
Individual
Self Awareness
Fogging
Frequency
Cleaning
Practices
Preventive
Practices
Predictive
Model
Classification
Recommend
MBD-based
information
Personalized E-Learning Environment
Self-Protection
Practices
Figure 1. A predictive system to classify the user’s misconception of MBDs
in a personalized e-learning environment
To curb the MBDs particularly in hotspot areas, it is important to ensure that members of the public
have adequate awareness through effective educational intervention. Social media usage such as Facebook
and Twitter were the preferred source of dengue information among the public [19]. Health stakeholders
of many countries often tweet and create a special posting to announce the epidemic outbreak of dengue in
certain areas. This includes how to control the endemic and to provide early information about disease
detection to the public. However, accessing such information might be hard for the people who lived in
the rural and remote areas. Relevant MBDs-based information can be recommended to mold positive
attitudes and educate the best preventive practices among the children. For instance, personal attributes and
environmental factors such as region, specific location features and climate contexts were used in [20]
to predict the human's risk of contracting infectious diseases. While in [21], an application is also made for
the user to submit the information related to mosquito prevalence (e.g. felt, seen, bitten, heard mosquitoes).
The public also can report the presence of Aedes that is a vector mosquito at the current location immediately
and report dengue cases in the household or surrounding communities. This open application requires active
citizen engagement and combined with other data, such as ovitrap egg counts and micro-climate data to
propose the stakeholders proactively in disease prevention, control and education. However, the application
might faces underreport when there is no participation from the people at certain places.
2. RESEARCH METHOD
Dengue knowledge, attitudes, and practices surveys have been conducted frequently to describe
the communities’ prevention towards the disease. It is expected that the knowledge of dengue and Aedes is
high among the public due to the good risk practices for dengue and good risk perception that have been
promoted through mainstream media and in education. An online community-based survey was carried out
on the 1st September 2018 until 10th December 2018 with 640 respondents of the Malaysian public aged
7 years old and above. The questionnaire is posted and shared in a few Malaysian Facebook science-based
pages and other groups.
The self-administered structured questionnaire covered all aspects of demographic profiles include
gender, age, marital status, education level, and housing type. Concerning the awareness of the dengue issue,
the participants need to answer general knowledge of mosquitoes such as the mosquito’s breeding sites,
mosquito’s prevention method and mosquito’s active biting time. Concerning to the awareness of dengue
presence and prevention practices, the participants were asked to provide information regarding the fogging
frequency in their living area. The fogging activity, in this context, was interpreted as their location might be
a hotspot area or not. This follows a section on thirteen questions in the True/False setup relevant to
cleanliness, trash management, water reservoir management and mosquito repellant usage. The logical flow

Bulletin of Electr Eng & Inf ISSN: 2302-9285 
Predicting the misconception of dengue disease based on… (Abrar Noor Akramin Kamarudin)
1633
of the questionnaire culminated as a part of the action by the study participants associated with the general
knowledge of mosquito behavior.
The online survey is the fastest way to collect data as respondents’ backgrounds are varied while
keeping their anonymity. The items on the questionnaire were adapted from previously validated instruments
with some modifications to meet the research objective. Close-ended questions are used to enhance
the simplicity and easiness of analysis. Respondents just need to choose the best answer according to their
existing knowledge.
Both descriptive analysis and machine learning were used in the experiment. Descriptive analysis
was utilized to summarize the socio-demographic such as age, gender, marital status, ethnicity, highest
education and type of residence. The machine learning algorithms were tested on 18 attributes of individual
knowledge, attitudes and practices such as cleanliness; waste management; clogged drains and stagnant water
handling; aerosol spray or mosquito repellent usage; cloth types and color usage, and the misconception
of mosquito’s breeding site, mosquito’s active bite time and dengue disease transmission. Table 1 described
the selected features that are found relevant to the proposed framework.
Table 1. Selected features mapping to the personalized e-learning environment
No. Features Importance
1. Aware of the Aedes mosquitoes presence within the living area To promote mosquito control practices through
educational programs where strategies related to
environmental management can be developed for
mosquito control when there is an appearance of it.
2. Fogging in housing area frequency To identify the availability of stagnant water that are
not disposed of properly or taken care of by the
communities as a potential breeding site.
3. Cleaning the area around the house To identify the practices of cleaning where everyone
lived in a different area and they might have a
different set of attitudes and practices of the cleaning
habit.
4. Dumping trash in the right place
5. Dumping trash that can hold water in the right place
6. Cleaning the drains or clogged waterways frequency
7. Closed or discarded water reservoir before a long vacation To identify the level of preventive practices where
different demographics might have a different type
of water management in their living space.
8. Checking larvae at the water reservoir
9. Putting Abate/chemicals in water reservoirs to avoid larvae
10. Wearing long-sleeved shirts and long pants to prevent mosquito bites To identify the level of self-protection practices as
the smallest effort that can be done to avoid
mosquito’s bite.
11. Using mosquito repellents such as liquids, coils, or trap in the living area
12. Using an aerosol spray to disperse mosquitoes in the dark area
13. Wearing bright colored clothes to avoid mosquito bites
14. Sleeping in the mosquito net
15. Applying mosquito repellent cream on the body
The machine learning algorithms training and testing were performed by using WEKA application,
a data mining application developed at the University of Waikato in New Zealand. Different kinds of
predictive models were built and the forecasting performances of those models were evaluated.
The algorithms used in the model building are support vector machine (SVM), decision tree (DT), Logistic
regression (LR), naïve bayes (NB) algorithm and artificial neural network (ANN). Tenfold cross-validation
technique was utilized to obtain the accuracy of each algorithm in classification. Such classification
algorithms comparison had been done to predict the chronic kidney disease (CKD) where naïve bayes
performs the highest accuracy [22].
3. RESULTS AND DISCUSSION
There are 640 respondents who administered the online survey. When the gender ratio is concerned,
the relative proportion of females respondents was higher (80%) when compared with male participants
Table 2. 76.4% of the respondents are aged 18 to 40 years old, 12.5% aged 7 to 17 years old and 11.1% aged
more than 40 years old where more than 60% of them are married. The largest number of the respondents is
coming from the Malay ethnic group (93.1%) followed by others (Chinese, Indian, Dusun, Kadazan,
Melanau, Bidayuh, Brunei, and Iban). Most of the respondents have attained tertiary education (73.5%)
followed by secondary school (26%) and primary school (0.5%). This illustrates that most respondents are
well educated and should have minimal knowledge of MBDs especially dengue fever. It is also found that
most of the respondents lived in a terrace or a twin house (43.7%).
Amongst the survey participants, 91.9% of respondents declare that they are aware of the presence
of Aedes mosquitoes in their area shown in Table 2. When asked about the frequency of fogging in their
residential area, more than 10% of respondents reported that it always happens in their place. Over 50%
of respondents described that their house has medium to many plants, and near 70% of them are living out
of the urban area.

 ISSN:2302-9285
1634
Table 2. Demographic information about participants surveyed
Characteristics N %
Sex
Female 512 80
Male 128 20
Age (range)
7-17 80 12.5
18-40 489 76.4
41+ 71 11.1
Marital Status
Married 409 63.9
Single 231 36.1
Ethnicity
Malay 591 93.1
Others 49 6.9
Highest education
Primary school 3 0.5
Secondary school 166 26
Tertiary education 470 73.5
Types of residence
Flat/Apartment/Condo 168 26.3
Terrace/Twin House 280 43.7
Bungalow/Village house 192 30
The majorities of people surveyed have never experienced dengue fever; however, near 20% of them
have gone through it before. It is found that 18% of respondents did not clean the area around their house and
there might be a correlation to the rise of the dengue fever. Furthermore, it is also reported that the prevention
practices among the respondents were very high, where 98.9% of them throw the trash in the right place.
It is found that more than 80% of respondents were disposing of the rubbish that can hold water in
the right place; ensure the water reservoir is closed or discarded before a long vacation, and; always check
the water reservoir to avoid larvae. Additionally, the number of other characteristics has varied where near
30% respondents have never cleaned the drains or clogged waterways; 28% of them did not use an aerosol
spray to disperse mosquitoes at dark area; near 40% respondents did not use mosquito repellent such as
liquids, coils, or trap in their living area; and only 25% respondents put abate or chemicals in water reservoirs
to avoid mosquito larvae. Furthermore, near half of respondents wear long-sleeved shirts and long pants to
prevent mosquito bites and wear bright-colored clothing to prevent mosquito bites. Only 3.9% of respondents
used a mosquito net for sleeping and 11.4% apply the mosquito repellent cream to their body.
Although 91.1% of respondents declared that they were aware of the Aedes mosquito’s appearance,
there were 67% of respondents who cannot answer the three questions correctly as illustrated in Table 3.
Thus, they are classified as having a misconception of the MBDs issue. The previous study shows the majority
of respondents just finished secondary school and only 54.6% of them have a high level of knowledge
regarding the dengue infections [15]. This result indicates that the current number of misconceptions is very
high among the public although they are highly educated. Respondents might think they have sufficient
knowledge regarding the dengue through the information they obtained from the television, newspaper, radio,
social media, and awareness campaign, but that might not comprehend or half delivered where more than half
of them are found unable to answer the questions correctly. Therefore, it is very important for the education
stakeholders to include the prediction system into their e-learning system to identify the target users and increase
their MBDs awareness in a personalized manner.
In order to correctly identify a person’s understanding regarding the MBDs issue, the performance
of selected machine learning algorithms was evaluated. The selected algorithms will be used for training
and testing to classify the person whether they need further intervention based on the 18 features as listed in
Tables 3 and 4. The true-positive (TP) rate, false-positive (FP) rate, precision, recall, F-measure, and time
took to build the classifier are presented in Table 5 for each algorithm. It is found that the SVM and DT
are the most precise algorithms to classify the person’s misunderstanding with the precision of 0.994
and 0.008 false-positive rate. However, DT took lesser time than SVM to classify due to the input type
was categorical and not continuous. Surprisingly, ANN performs inferior to the logistic regression model as
the model-building process is less complicated for logistic regression, and may be considered not worthy
of the artificial neural networks. In fact, the ANN model can be seen as nonlinear generalizations of logistic
regression, and thus at least as powerful as that model which in this case, it was not significantly better to use
neural networks.
Based on the result, the DT algorithm is found suitable to be used in the predictive model to support
the objective of the research. Traditional dengue fever classification by using Naïve Bayes algorithm has

1635
been conducted in [23] to identify positive or negative results, based on several attributes such as fever,
bleeding, flu, myalgia, and other symptoms with the precision of 0.92. As in [24], the DT algorithm has been
used to predict patients with severe dengue that may warrant admission based on several variables
(e.g. vomiting, pleural effusion, and systolic blood pressure) which return precision rate up to 0.81. It is also has
been used in [25] to propose a patient-monitoring plan and outpatient management of fever in resource-poor
settings by using the clinical features and laboratory indicators to identify the severity of the dengue cases.
A study suggests the increase of knowledge through social media can influence the positive attitude
and practices of dengue prevention measures [26]. However, social media alone cannot be personalized to
fulfill the criteria as human attitude regarding the MBDs was not taken into consideration. This study is
proposing the misconception of the disease can be identified among the public in the form of a quiz upon
using the e-learning system. Once the system obtained enough information regarding the user’s awareness
background, it will analyze and personalized the relevant information regarding the MBDs between
the learning sessions. The information is also can be fed directly to the target user, such as the children.
His can be done based on a few rules where the children might be at a very young age of using the Internet or
the children who lived in rural areas. Otherwise, feeding such information to a non-target user will be
wasting the resources. Further investigation is needed to identify the differences of culture, socioeconomic
background, and location which may also contribute to the preventive practices among the children.
Thus, this paper highlights the big opportunity for the health authorities and education stakeholders to
promote such campaigns through e-learning systems or any available open learning platform.
Table 3. Awareness of dengue presence and prevention practices
No. Features N %
1.
I am aware of the presence of Aedes mosquitoes in my area
Yes
No
588
52
91.9
8.1
2.
What is the frequency of fogging in your housing area?
Always
Sometimes
Seldom
None
65
173
280
122
10.2
27
43.8
19.1
3.
I often clean the area around the house
Yes
No
525
115
82
18
4.
I dump trash in the right place
Yes
No
633
7
98.9
1.1
5.
I dump trash that can hold water in the right place
Yes
No
559
81
87.3
12.7
6.
I often clean the drains or clogged waterways
Yes
No
460
180
71.9
28.1
7.
I make sure the water reservoir is closed or discarded before a long vacation
Yes
No
552
88
86.3
13.7
8.
I always use an aerosol spray to disperse mosquitoes in the dark area
Yes
No
461
179
72
28
9.
I always check the water reservoir so there are no larvae
Yes
No
547
93
85.5
14.5
10.
I use mosquito repellents such as liquids, coils, or trap in my living area
Yes
No
395
245
61.7
38.3
11.
I put Abate/chemicals in water reservoirs to avoid larvae
Yes
No
161
479
74.8
25.2
12.
I wear long-sleeved shirts and long pants to prevent mosquito bites
Yes
No
324
316
50.6
49.4
13.
I wear bright colored clothes to avoid mosquito bites
Yes
No
310
330
48.4
51.6
14.
I sleep in the mosquito net
Yes
No
25
615
3.9
96.1
15.
I always apply mosquito repellent cream on my body
Yes
No
73
567
11.4
88.6

 ISSN:2302-9285
1636
Table 4. Knowledge of Aedes characteristics
No. Questions N %
1.
I know Aedes mosquito breeding in the following areas:
Stagnant water
Septic tank
Drain with waste
Container with clear water
2.
I know Aedes mosquitoes actively bites at the following times:
Dawn
Afternoon
Evening
Night
3.
How is dengue transmitted?
Mosquito bites
Contaminated and exposed foods
Don’t know
Correctly answer all questions 211 33
Incorrectly answer some/all questions 429 67
Table 5. Performance comparison of the classification algorithms
Algorithms TP rate FP rate Precision Recall F-measure Time taken (second)
NB 0.991 0.009 0.991 0.991 0.991 0.01
LR 0.991 0.012 0.991 0.991 0.991 0.24
ANN 0.986 0.012 0.986 0.986 0.986 6.32
SVM 0.994 0.008 0.994 0.994 0.994 0.16
DT 0.994 0.008 0.994 0.994 0.994 0.05
4. CONCLUSION
Based on the awareness survey, the majority of respondents applied good practices in dengue
preventive measures. An effective prediction model to detect the MBDs misconception is still needed to
identify the communities’ awareness and enhance their knowledge by recommending information about
dengue prevention and early symptom recognition. A predictive system should be embedded to analyze
the user’s behavior and provide personalized information. Several features that have been selected for
the study are found relevant for the proposed framework. This research proposes the Decision Tree algorithm
in the predictive model construction of the online dengue awareness program by taking into account
the human behavior and their current knowledge regarding the MBDs. The research focuses on Aedes
and dengue transmission due to the high number of dengue fever and dengue hemorrhagic fever cases in Malaysia
compared to other MBDs. Considerably more work is needed to identify features and causes of other MBDs such
as Malaria, Zika, and Chikungunya including geospatial analysis and environmental factors.
ACKNOWLEDGEMENTS
The first author would like to acknowledge the USM Fellowship Scheme for providing a scholarship
to conduct the research.
REFERENCES
[1] R. Jain, S. Sontisirikit, S. Iamsirithaworn, and H. Prendinger, “Prediction of dengue outbreaks based on disease
surveillance, meteorological and socio-economic data,” BMC infectious diseases, vol. 19, no. 1, pp. 1-16, 2019.
[2] P. E. Parham et al., “Climate, environmental and socio-economic change: Weighing up the balance in vector-borne
disease transmission,” Philosophical Trans. of the Royal Soc. B: Biological Sci., vol. 370, no. 1665, pp. 1-17, 2015.
[3] W. J. Tabachnick, “Challenges in predicting climate and environmental effects on vector-borne disease episystems
in a changing world,” Journal of Experimental Biology, vol. 213, no. 6, pp. 946-954, 2010.
[4] S. Deb, C. M. L. Acebedo, G. Dhanapal, and C. M. C. Heng, “An ensemble prediction approach to weekly Dengue
cases forecasting based on climatic and terrain conditions,” J. of Health and Soc. Sci., vol. 2, no. 3, pp. 257-272, 2017.
[5] R. Zaki et al., “Public perception and attitude towards dengue prevention activity and response to dengue early
warning in Malaysia,” PLoS One, vol. 14, no. 2, pp. 1-22, 2019.
[6] M. K. Othman and M. S. N. M. Danuri, “Proposed conceptual framework of dengue active surveillance system (DASS) in
Malaysia,” 2016 International Conference on Information and Communication Technology, pp. 90-96, 2017.
[7] B. Modu, N. Polovina, Y. Lan, S. Konur, A. Asyhari, and Y. Peng, “Towards a predictive analytics-based
intelligent malaria outbreak warning system,” Applied Sciences, vol. 7, no. 8, pp. 1-20, 2017.
[8] H. L. Lee et al., “Dengue vector control in Malaysia-challenges and recent advances,” IIUM Medical Journal
Malaysia, vol. 14, no. 1, pp. 11-16, 2015.

1637
[9] J. K. Davis, G. Vincent, M. B. Hildreth, L. Kightlinger, C. Carlson, and M. C. Wimberly, “Integrating environmental
monitoring and mosquito surveillance to predict vector-borne disease: Prospective forecasts of a west nile virus outbreak,”
PLoS Currents, vol. 9, 2017.
[10] A. L. V. Sharma, A. Kumar, L. Panat, and G. Karajkhede, “Malaria outbreak prediction model using machine learning,”
International Journal of Advanced Research in Computer Engineering & Technology, vol. 4, no. 12, pp. 4415-4419, 2015.
[11] K. Kesorn et al., “Morbidity rate prediction of dengue hemorrhagic fever (DHF) Using the support vector machine
and the aedes aegypti infection rate in similar climates and geographical areas,” PLoS One, vol. 10, no. 5, 2015.
[12] A. Itrat et al., “Knowledge, awareness and practices regarding dengue fever among the adult population of dengue
hit cosmopolitan,” PloS one, vol. 3, no. 7, pp. 1-6, 2008.
[13] M. M. T. Aung et al., “Knowledge, attitude, practices related to dengue fever among rural population in
Terengganu, Malaysia,” Malaysian Journal of Public Health Medicine, vol. 16, no. 2, pp. 15-23, 2016.
[14] S. Boonchutima, K. Kachentawa, M. Limpavithayakul, and A. Prachansri, “Longitudinal study of Thai people
media exposure, knowledge, and behavior on dengue fever prevention and control,” Journal of Infection and Public
Health, vol. 10, no. 6, pp. 836-841, 2017.
[15] F. M. Yussof, A. Hassan, T. Zin, T. M. A. R. Hussin, N. Kadarman, and R. Umar, “Knowledge of dengue among students
in universiti sultan zainal abidin (Unisza), Terengganu, Malaysia and the Influence of knowledge of dengue on attitude
and practice,” Journal of Fundamental and Applied Sciences, vol. 2017, no. 2S, pp. 199-216, 2017.
[16] S. Jeelani, S. Sabesan, and S. Subramanian, “Community knowledge, awareness and preventive practices regarding
dengue fever in Puducherry-South India,” Public Health, vol. 129, no. 6, pp. 790-796, 2015.
[17] M. M. Mathania, S. I. Kimera, and R. S. Silayo, “Knowledge and awareness of malaria and mosquito biting behaviour in
selected sites within Morogoro and Dodoma regions Tanzania,” Malaria journal, vol. 15, no. 1, pp. 1-9, 2016.
[18] M. Dhimal et al., “Knowledge, attitude and practice regarding dengue fever among the healthy population
of highland and lowland communities in Central Nepal,” PLoS One, vol. 9, no. 7, pp. 1-15, 2014.
[19] W. R. W. Rosli, S. A. Rahman, J. K. Parhar, and M. I. Suhaimi, “Positive impact of educational intervention on
knowledge, attitude, and practice towards dengue among university students in Malaysia,” Journal of Public
Health, vol. 27, no. 4, pp. 461-471, 2019.
[20] R. A. Vinarti and L. M. Hederman, “A personalized infectious disease risk prediction system,” Expert Systems with
Applications, vol. 131, pp. 266-274, 2019.
[21] A. N. Babu et al., “Smartphone geospatial apps for dengue control, prevention, prediction, and education: MOSapp,
DISapp, and the mosquito perception index (MPI),” Envi. Monitoring and Assessment, vol. 191, no. 2, p. 393, 2019.
[22] Z. Saringat, A. Mustapha, R. D. R. Saedudin, and A. A. Samsudin, “Comparative analysis of classification
algorithms for chronic kidney disease diagnosis,” Bulletin of Elec. Eng. and Inf., vol. 8, no. 4, pp. 1496-1501, 2019.
[23] K. Shaukat Dar and S. M. Ulya Azmeen, “Dengue fever prediction: A data mining problem,” Journal of Data
Mining in Genomics & Proteomics, vol. 06, no. 3, pp. 1-5, 2015.
[24] J. Tamibmaniam, N. Hussin, W. K. Cheah, K. S. Ng, and P. Muninathan, “Proposal of a clinical decision tree
algorithm using factors associated with severe dengue infection,” PLoS One, vol. 11, no. 8, pp. 1-10, 2016.
[25] K. Phakhounthong et al., “Predicting the severity of dengue fever in children on admission based on clinical features
and laboratory indicators: Application of classification tree analysis,” BMC pediatrics, vol. 18, no. 1, pp. 1-9, 2018.
[26] W. R. Wan Rosli, S. Abdul Rahman, J. K. Parhar, and M. I. Suhaimi, “Positive impact of educational intervention on
knowledge, attitude, and practice towards dengue among university students in Malaysia,” Journal of Public Health, 2018.
BIOGRAPHIES OF AUTHORS
Abrar Noor Akramin Kamarudin is a Ph.D. student at the School of Computer Science,
Universiti Sains Malaysia. His research is focused on vector-borne disease prediction as well as
personalizing the education for the community.
Zurinahni Zainol is an Associate Professor at the School of Computer Science, Universiti Sains
Malaysia. Her research specialization is Theory of Database Design and Management, XML
Document Schema Design, Data Model, and Software Development using Formal Specification.
Nur Faeza Abu Kassim is an Associate Professor at the School of Biological Science,
Universiti Sains Malaysia. She is an expert in medical and epidemiological entomology. Her
research specialization is mosquito-borne disease & the role of vectors in transmission disease.
Development of the integrated vector management, mosquito control monitoring/surveillance
program, attractant and trapping system.

Predicting the misconception of dengue disease based on the awareness survey

More Related Content

What's hot (11)

Similar to Predicting the misconception of dengue disease based on the awareness survey (20)

More from journalBEEI (20)

Recently uploaded (20)

Predicting the misconception of dengue disease based on the awareness survey