IMPELEMENTASI SEMI-SUPERVISED LEARNING PADA PERSONALIZED ASTHMA MANAGEMENT SYSTEM

Cut Fiarni, Evasaria Magdalena Sipayung, Kevin Barry Moningka

Abstract


Asthma is a chronic disease of the lungs that react on various stimuli that was found on the patient’s body. The various stimuli, which are Sensitization and Inflammatory are different   for each patient and it also could lead to asthma attack on different severity degrees. Information and knowledge regarding the cause factors of asthma is very important for patient, so they could have a better control of their asthma triggers and health conditions.  In this paper, we developed a personalized asthma management system by using semi-supervised learning technique.  The main methodology is to find pattern from patient daily information, then system will extract rules regarding their asthma trigger and classify them to each of asthma severity degrees. There is a dashboard that contain all the factors noted by the patient and evaluation of their asthma management. The result of experiment evaluation shown that the proposed system have 80% of accuracy, which proves that system reliable for a better asthma self-management.

Keywords: asthma management,semi- supervised learning, dashboard system

Penyakit asma adalah penyakit radang kronis pada paru-paru yang bereaksi pada berbagai rangsangan yang terdapat pada tubuh penderitanya. Penyakit ini memiliki berbagai macam faktor penyebab terjadinya serangan asma dan tidak dapat digeneralisasikan. Selain itu, penyakit asma memiliki derajatnya masing-masing sesuai dengan tingkat keparahannya. Informasi tentang faktor penyebab asma penting karena penderita asma cenderung lalai dalam memperhatikan gejala atau faktor-faktor penyebab terjadinya asma sehingga mengakibatkan manajemen asma yang tidak baik. Pada penelitian ini dikembangkan aplikasi manajemen penyakit asma yang bersifat personal untuk masing-masing penderita, dengan mengadopsi teknik supervised learning. Data dan infromasi aktivitas harian penderita akan direkam oleh system, kemudian system akan mencari pola terkait faktor-faktor pemicu dan pemacu asma, serta mengklasifikasikannya berdasarkan pada  derajat serangan asma. Pada system usulan terdapat dashboard yang memberikan informasi dan hasil evaluasi kondisi historis penderita asma secara mudah dan efektif. Dari hasil pengujian  didapat akurasi system sebesar 80%, hal ini menunjukan system mampu membantu pasien dalam melakukan manajemen asma secara mandiri.

Kata kunci: manajemen penyakit asma, semi-supervised learning, dashboard system


Full Text:

PDF

References


D. Belgrave, J. Henderson, A. Custovic., “Disaggergating asthma : Big investigation versus big data”, Journal of Allergy and Clinical Immunology, ScienceDirect, UK, 2016.

Fiarni, Cut., 2014. “Design of Personalized Asthma Management with Data Mining Methods”, Proceeding of International on Electrical Engineering, Computer Science and Informatics (EECSI), Yogyakarta, p120-123, 2014.

H.T Chu, C.C.Huang, Z.H.Liab,J.J.P. Tsai, “A ubiquitous warning system for asthma-inducement,” IEEE Sensor Networks, Ubiquitous, and Trustworthy Computing, 2006

Do,Q., Robinson, K., Tran, S, “Big data analysis:Why not an asthma app?,” Proceeding of Informing Science & IT Education Conference (inSITE),p p155-170, 2015.




DOI: http://dx.doi.org/10.20527/klik.v4i1.69

Copyright (c) 2017 KLIK - KUMPULAN JURNAL ILMU KOMPUTER

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.joomla
counter View My Stats