Image of PERBANDINGAN METODE K-NEAREST NEIGHBOUR DAN FUZZY K-NEAREST NEIGHBOUR DALAM KLASIFIKASI DATA

Skripsi

PERBANDINGAN METODE K-NEAREST NEIGHBOUR DAN FUZZY K-NEAREST NEIGHBOUR DALAM KLASIFIKASI DATA

Penilaian

0,0

dari 5
Penilaian anda saat ini :  

K-NN is such an effective and the most commonly used method for data classification. This method also believed as one of the most popular and easiest method to be implemented in this subject. But, there is also a method that has a lot of similarities with this one ( K-NN ) and known for its ability to determine an object category also by its weight, this method called Fuzzy K-NN. Fuzzy K-Nearest Neighbour basically is just the K-NN with Fuzzy theory in it. The only difference between this two is the ability of Fuzzy K-NN to determines the membership value in each individuals in data classification process. in this study, writer tried to examine the affects of weights ( leads to membership value ) on the accuracy of the data classification ( in each K ) by using iris and blood transfusion data as the objects. Based on the research results, it was found that the best accuracy for both K-NN and Fuzzy K-NN are 98,04% for iris data, with K = 1 in K-NN and K = 13 in Fuzzy K-NN. Meanwhile, in blood transfusion data classification the best results for both methods shows that K-NN has a better accuracy ( 80,54% ) than the Fuzzy K-NN ( 78,95% ) in K = 20. Keywords : Classification, K-nearest neighbour, Fuzzy K-nearest neighbour, weights, Membership value

Availability
Inventory Code Barcode Call Number Location Status
2007000069T41368T413682020Central Library (REFERENSI)Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T413682020
Publisher
Inderalaya : Jurusan Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya.,
Collation
xviii, VI-2 hlm, 29 cm
Language
Indonesia
ISBN/ISSN
-
Classification
005.120 7
Content Type
Text
Media Type
other
Carrier Type
-
Edition
-
Subject(s)
Jurusan Teknik Informatika
Pemograman Aplikasi
Specific Detail Info
-
Statement of Responsibility
Other version/related

No other version available

File Attachment
Comments
You must be logged in to post a comment