Implementasi Algoritma Modified K-Nearest Neighbor (MKNN) Untuk Klasifikasi Penyakit Demam

Wafiyah, Fakihatin (2017) Implementasi Algoritma Modified K-Nearest Neighbor (MKNN) Untuk Klasifikasi Penyakit Demam. Sarjana thesis, Universitas Brawijaya.

Abstract

Demam merupakan indikator awal untuk beberapa penyakit antara lain demam berdarah, tifoid dan malaria disertai gejala yang mirip, antara lain nyeri otot, gangguan pencernaan, kondisi lidah serta pembesaran pada hati dan limpa. Kemiripan gejala dari masing-masing penyakit sering menimbulkan kesulitan dalam mendapatkan anamnese (diagnosa sementara) sehingga pasien mendapatkan penanganan awal yang kurang tepat. Untuk menangani masalah tersebut diperlukan teknologi untuk mendapatkan diagnosa sementara dengan menerapkan salah satu metode klasifikasi yaitu Modified K-Nearest Neighbor (MKNN). Metode tersebut mempelajari pola dari data hasil pemeriksaan sebelumnya berdasarkan 15 gejala penyakit dengan proses perhitungan jarak eucledian, perhitungan nilai validitas dan perhitungan weighted voting yang hasil akhirnya digunakan untuk penetapan kelas klasifikasi berdasarkan nilai K yang telah ditentukan. Berdasarkan hasil pengujian terhadap perubahan nilai K, perubahan jumlah data latih dan perubahan komposisi data latih didapatkan rata-rata akurasi untuk pengujian pengaruh nilai K terhadap akurasi sebesar 88.55%. Nilai rata-rata akurasi yang didapatkan dari pengujian pengaruh variasi jumlah data latih adalah 92.42%. Pengujian pengaruh komposisi data latih terhadap akurasi mendapatkan nilai rata-rata akurasi sebesar 87.89%. Pengujian pengaruh komposisi data latih dan data uji terhadap akurasi mendapatkan nilai rata-rata akurasi sebesar 96.35%.

English Abstract

Fever is an early indicator for some diseases such as dengue fever, typhoid and malaria accompanied by similar symptoms, including muscle pain, indigestion, tongue condition and enlargement of the liver and spleen. Similar symptoms of each disease cause difficulties in getting anamnese (temporary diagnosis) so that patients get the inadequate initial treatment. Handling the problem, technology is needed to obtain a temporary diagnosis by applying one of the classification method of Modified K-Nearest Neighbor (MKNN). The method studied the pattern of previous examination data based on 15 symptoms of disease with eucledian distance calculation process, calculation of validity value and weighted voting calculation that the end result is used for class classification determination based on predetermined value of K. Testing of the value of K get the accuracy of 88.55%. The average value of accuracy obtained from testing of variation in the amount of training data is 92.42%. Testing the influence of the composition of train data get the average value of accuracy of 87.89%. Testing the influence of the composition of train data and test data get the average value of accuracy of 96.35%.

Item Type: Thesis (Sarjana)
Identification Number: SKR/FTIK/2017/288/051705328
Uncontrolled Keywords: Modified K-Nearest Neighbor (MKNN), Klasifikasi, Demam, Malaria, Tifoid, Demam Berdarah (DBD)
Subjects: 000 Computer science, information and general works > 004 Computer science > 004.015 1 Finite mathematic
Divisions: Fakultas Ilmu Komputer > Teknik Informatika
Depositing User: Yusuf Dwi N.
Date Deposited: 28 Jul 2017 02:28
Last Modified: 21 Nov 2020 11:21
URI: http://repository.ub.ac.id/id/eprint/757
[thumbnail of FAKIHATIN WAFIYAH .pdf] Text
FAKIHATIN WAFIYAH .pdf

Download (6MB)
[thumbnail of Thumbnails conversion from text to thumbnail_lightbox] Other (Thumbnails conversion from text to thumbnail_lightbox)
lightbox.jpg

Download (68kB)
[thumbnail of Thumbnails conversion from text to thumbnail_preview] Other (Thumbnails conversion from text to thumbnail_preview)
preview.jpg

Download (25kB)
[thumbnail of Thumbnails conversion from text to thumbnail_medium] Other (Thumbnails conversion from text to thumbnail_medium)
medium.jpg

Download (7kB)
[thumbnail of Thumbnails conversion from text to thumbnail_small] Other (Thumbnails conversion from text to thumbnail_small)
small.jpg

Download (2kB)

Actions (login required)

View Item View Item