Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Katrina Puspita
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Klasifikasi Aktivitas Manusia menggunakan Algoritme Fuzzy Learning Vector Quantization (FLVQ) dengan Reduksi Dimensi Principal Component Analysis (PCA) Katrina Puspita; Fitra Abdurrachman Bachtiar; Novanto Yudistira
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 6 No 1 (2022): Januari 2022
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Human Activity Recognition is a popular research topic which aims to identify human activities using sets of data obtained with the help of sensor or image recorder. Research results regarding human activity recognition are widely used for military, health, and security purposes. Previously, research has been conducted regarding human activity recognition using Learning vector quantization (LVQ), but this method has several weaknesses which are LVQ accuracy rate rely heavily on the pre-processing of the data, sensitive to overlapping dataset, and requires a long computational time, in which these problems affect the performance of LVQ algorithm. One of the methods used to solve these problems is Fuzzy learning vector quantization (FLVQ) with dimensionality reduction Principal Component Analysis (PCA). FLVQ algorithm is a development from LVQ algorithm, where it uses batch of Fuzzy C-Means and LVQ in its calculations. The research was conducted using various tests on the parameters and the number of classes to be used in the classification stage. The highest accuracy was obtained from the classification using 2 classes, which gained 52.11% in accuracy with 30 features out of the total features (561 features) being used.
Co-Authors Fitra Abdurrachman Bachtiar Novanto Yudistira