Tumangger, Ricky Marten Sahalatua (2020) Komparasi Metode Data Mining Support Vector Machine Dengan Naive Bayes Untuk Klasifikasi Status Kualitas Air. Sarjana thesis, Universitas Brawijaya.
Abstract
Air merupakan senyawa kimia yang sangat berperan penting bagi setiap makhluk hidup untuk kelangsungan hidup. Dibumi air memiliki wilayah yang sangat luas dibandingkan dengan daratan. Senyawa air mempunyai wilayah hampir 71% mengelilingi daratan. Air tersebut juga terdiri dari air laut, sungai, danau, air tanah, air rawa, salju dan uap yang berada di lapisan udara dan mengandung zat-zat mineral yang terlarut dalam air. Untuk menetukan klasifikasi status kualitas air menggunakan metode Support Vector Machine dan Naive Bayes. Metode ini dipilih karena penelitian sebelumnya mendapatkan hasil akurasi yang tinggi untuk pengklasifikasian. Adapun parameter-parameter yang digunakan adalah derajat keasaman (pH), TDS, NO2, NO3, kesadahan, khlorida, mangan. Metode Suppord Vector Machine dan Naive Bayes akan memberikan keluaran hasil perbandingan akurasi antara kedua metode tersebut. Pengujian pada sistem dilakukan dengan menggunakan pengujian K-Fold Cross Valadation dengan mendapatakan hasil akurasi tertinggi ketika K=9 untuk metode Support Vector Machine dan K=5 untuk metode Naive Bayes. Pengujian parameter untuk metode Support Vector Machine mendapatkan akurasi tertinggi ketika nilai thresholdnya 10−9, C=3,
English Abstract
Water is a chemical compound that is very important for every living thing for survival on this earth. On earth water has a very large area compared to the mainland this compound has an area of almost 71% around the land. The water also consists of sea water, rivers, lakes, ground water, swamp water, snow and steam which are in the air layer that contains mineral substances that are recruited in water. To determine the classification of water quality status using the Support Vector Machine and Naive Bayes methods. This method was chosen because previous studies get high accuracy results for classification. The parameters used are the degree of acidity (pH), TDS, NO2, NO3, hardness, chloride, manganese. The Vector Machine and Naive Bayes Suppord method will provide the results of the comparison of the accuracy of the two methods. Testing on the system is done using the K-Fold Cross Valadation test with the highest accuracy results when K = 9 for the Support Vector Machine method and K = 5 for the Naive Bayes method. Testing parameters for the Support Vector Machine method gets the highest accuracy when the threshold value is 10−9, C = 3, γ = 0.01, λ = 2.5, maximum iteration value = 1, σ = 0.1. From these tests the accuracy of the Support Vector Machine method was 78.70% and the Naive Bayes method was 85.78%. The best results obtained by the classification of water quality status are the Naive Bayes method compared to using the Support Vector Machine method because the average accuracy of the Naive Bayes method is higher.
Other obstract
-
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/FILKOM/2020/111/052003073 |
| Uncontrolled Keywords: | kualitas air, Support Vector Machine, Naive Bayes, K-Fold Cross Validation, Threshold, C, |
| Subjects: | 000 Computer science, information and general works > 006 Special computer methods > 006.3 Artificial intelligence > 006.31 Machine learning > 006.312 Data mining |
| Divisions: | Fakultas Ilmu Komputer > Teknik Informatika |
| Depositing User: | Budi Wahyono Wahyono |
| Date Deposited: | 01 Aug 2020 08:57 |
| Last Modified: | 14 Apr 2023 01:41 |
| URI: | http://repository.ub.ac.id/id/eprint/180871 |
![[thumbnail of DALAM MASA EMBARGO]](http://repository.ub.ac.id/style/images/fileicons/text.png) |
Text (DALAM MASA EMBARGO)
Ricky Marten Sahalatua Tumangger.pdf Restricted to Registered users only until 31 December 2023. Download (5MB) |
Actions (login required)
 |
View Item |