Firdausanti, Neni Alya (2017) Klasifikasi Kelas Risiko Paien Pneumonia Menggunakan Metode Hybrid Analisis Diskriminan Linier-Particle Swarm Optimization (ADL-PSO) dan Naive Bayes Classification. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 1313100112-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
1313100112-Undergraduate_Thesis.pdf - Published Version Download (1MB) |
Abstract
sehingga penyebaran penyakit ini tergolong sangat cepat. Oleh karena itu diagnosa yang cepat dan tepat sangat diperlukan agar dapat menentukan penanganan dan perawatan yang tepat. Beberapa penilaian kelas risiko pneumonia dikembangkan untuk mempermudah diagnosis pneumonia. Terdapat berbagai sistem skoring PSI, CURB-65, modified ATS (m-ATS) dan sebagainya untuk menentukan kelas risiko pasien pneumonia, akan tetapi tidak ada patokan pasti untuk menentukan sistem apa yang harus digunakan untuk mengelompokan kelas risiko pneumonia. Oleh karena itu diperlukan studi klasifikasi untuk mengkaji variabel-variabel yang diapat digunakan untuk mengelompokan kelas risiko pneumonia secara tepat. Penelitian ini menerapkan metode Analisis Diskriminan Linier dengan seleksi variabel forward selection, backward elimination dan stepwise method, Hybrid Analisis Diskriminan Linier-Particle Swarm Optimization (ADL-PSO) dan Naïve Bayes untuk mengklasifikasikan kelompok kelas risiko pneumonia berdasarkan data rekam medis pasien kemudian ketiga metode klasifikasi tersebut dibandingkan nilai akurasinya. Hasil penelitian menunjukan bahwa metode klasifikasi terbaik adalah ADL-PSO.
=========================================================================================
Pneuomonia is a disease that is transmitted through the air so that the spread of this disease is very fast. Therefore a fast and precise diagnosis is necessary in order to determine appropriate treatment and care. Several scoring assessments of pneumonia were developed to facilitate the diagnosis of pneumonia. There are PSI scoring systems, CURB-65, modified ATS (m-ATS) and so on to determine the risk class of pneumonia patients, but there is no definite benchmark to determine what system should be used to classify the risk class of pneumonia. Therefore, a classification study is needed to assess the variables used to correctly classify the risk of pneumonia. This research applies Linear Discriminant Analysis method with selection of forward selection, backward elimination and stepwise method, Hybrid Linear Discriminant Analysis-Particle Swarm Optimization (LDA-PSO) and Naïve Bayes to classify pneumonia risk class group based on patient medical record data then all classification methods are compared the value of its accuracy. The results showed that the best classification method is LDA-PSO.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSSt 519.535 Fir k |
| Uncontrolled Keywords: | Analisis Diskriminan Linier, Naïve Bayes, Particle Swarm Optimization, Pneumonia, Seleksi Variabel |
| Subjects: | H Social Sciences > HA Statistics |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Neni Alya Firdausanti |
| Date Deposited: | 25 Oct 2017 07:26 |
| Last Modified: | 05 Mar 2019 08:38 |
| URI: | http://repository.its.ac.id/id/eprint/47887 |
Actions (login required)
 |
View Item |