Image of PERBANDINGAN PERFORMA METODE KNN DAN LMKNN TERHADAP KLASIFIKASI IRIS DENGAN VARIASI JUMLAH ATRIBUT DATA

Skripsi

PERBANDINGAN PERFORMA METODE KNN DAN LMKNN TERHADAP KLASIFIKASI IRIS DENGAN VARIASI JUMLAH ATRIBUT DATA

Penilaian

0,0

dari 5
Penilaian anda saat ini :  

The KNN (K-Nearest Neighbor) method is a classification technique that relies on concepts between data to predict new data classes. KNN has a variation of the method namely, LMKNN (Local Mean Based K-Nearest Neighbor) by using the average of several classes for more accurate predictions. To identify iris flowers based on their characteristics, namely, the length and width of the petals and the length and width of the crown, an accurate method is needed. Therefore, this study aims to evaluate and analyze the two methods in classifying iris flowers. Testing iris data with 3 attributes and iris2d data with 2 attributes, which both amount to 150 data with 3 classes. Tested using cross validation with 10 fold and confusion matrix to determine the performance value. In this study KNN produces the highest accuracy value of 97%, while the LMKNN algorithm gets the highest accuracy of 96%. The use of different data variations and nearest neighbor values did not significantly affect the performance of the two methods. Found data that has 3 attributes, getting the highest accuracy of 97%. Then the highest result of the nearest neighbor is k = 5 in the KNN method on Iris data, getting 95% accuracy, 98% precision, 97% Recall, 97% F1 Score.

Availability
Inventory Code Barcode Call Number Location Status
2307003697T127427T1274272023Central Library (Referens)Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T1274272023
Publisher
Inderalaya : Jurusan Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya.,
Collation
xiv, 56 hlm.; Ilus.; 29 cm
Language
Indonesia
ISBN/ISSN
-
Classification
004.07
Content Type
Text
Media Type
unmediated
Carrier Type
-
Edition
-
Subject(s)
Pemrosesan data
Jurusan Teknik Informatika
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