• Login
    View Item 
    •   USU-IR Home
    • Faculty of Computer Science and Information Technology
    • Department of Information Technology
    • Master Theses
    • View Item
    •   USU-IR Home
    • Faculty of Computer Science and Information Technology
    • Department of Information Technology
    • Master Theses
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Optimisasi Metode K-NN (K-Nearest Neighbour) Menggunakan Fuzzy Logic pada Klasifikasi Data

    View/Open
    Fulltext (907.9Kb)
    Date
    2021
    Author
    Harahap, Zulfadhli
    Advisor(s)
    Tulus
    Lydia, Maya Silvi
    Metadata
    Show full item record
    Abstract
    Classifications data can be undertaken by using a method of K-Nearest Neighbour through proximity the distance data training with the data that are being tested. A problem was often the case in the process of data processing using a method of K-Nearest Neighbour is the result the value of being ambiguous and it is not clear if the distance between data is too near. A new method is required in solving the problem. Fuzzy logic was used in the study in the methods of k-nearest neighbour to group its output target. The data used in this research was classifications a disease of the teeth to determine illnesses that fit on the teeth and classifications the type of leaves to determination of the types of what is fitting and becoming leaves on the leaves. A method of testing shows K-Nearest Neighbour and data obtained the level of accuracy of a disease of the teeth by 86 % and methods K-Nearest Neighbour with the data leaves obtained the level of accuracy of as much as 73.3 %. A method of testing shows K-Nearest Neighbour Fuzzy and data obtained the level of accuracy of a disease of the teeth of 93 % and methods K-Nearest Neighbour Fuzzy with the data leaves obtained the level of accuracy of 93 %. K-Nearest Neighbour Fuzzy prove even more reliable methods used in data processing to classification.
     
    Klasifikasi data dapat dilakukan dengan menggunakan metode K-Nearest Neighbour melalui kedekatan jarak data latih dengan data yang diuji. Masalah yang sering terjadi dalam proses pengolahan data menggunakan metode K- Nearest Neighbour adalah hasil nilai menjadi ambigu dan dan tidak jelas jika jarak antar data terlalu dekat. Diperlukan metode baru dalam memecahkan masalah tersebut. Penelitian ini menggunakan Fuzzy logic dalam metode K-Nearest Neighbour untuk mengelompokkan target output. Data yang digunakan dalam penelitian ini adalah klasifikasi penyakit gigi untuk menentukan jenis penyakit yang sesuai pada gigi dan klasifikasi jenis daun untuk menentukan jenis daun apa yang sesuai pada daun. Hasil uji metode K-Nearest Neighbour dengan data penyakit gigi diperoleh tingkat akurasi sebesar 86% dan metode K-Nearest Neighbour dengan data daun diperoleh tingkat akurasi sebesar 73.3%. Hasil uji metode K-Nearest Neighbour Fuzzy dan data penyakit gigi diperoleh tingkat akurasi sebesar 93% dan metode K- Nearest Neighbour Fuzzy dengan data daun diperoleh tingkat akurasi sebesar 93%. Metode K-Nearest Neighbour Fuzzy terbukti lebih reliable untuk digunakan dalam pengolahan data klasifikasi.

    URI
    https://repositori.usu.ac.id/handle/123456789/49841
    Collections
    • Master Theses [621]

    Repositori Institusi Universitas Sumatera Utara (RI-USU)
    Universitas Sumatera Utara | Perpustakaan | Resource Guide | Katalog Perpustakaan
    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV
     

     

    Browse

    All of USU-IRCommunities & CollectionsBy Issue DateTitlesAuthorsAdvisorsKeywordsTypesBy Submit DateThis CollectionBy Issue DateTitlesAuthorsAdvisorsKeywordsTypesBy Submit Date

    My Account

    LoginRegister

    Repositori Institusi Universitas Sumatera Utara (RI-USU)
    Universitas Sumatera Utara | Perpustakaan | Resource Guide | Katalog Perpustakaan
    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV