SIMILARITY BASED ENTROPY ON FEATURE SELECTION FOR HIGH DIMENSIONAL DATA CLASSIFICATION

Abstract: Curse of dimensionality is a major problem in most classification tasks. Feature transformation and feature selection as a feature reduction method can be applied to overcome this problem. Despite of its good performance, feature transformation is not easily interpretable because the physical meaning of the original features cannot be retrieved. On the other side, feature selection with its simple computational process is able to reduce unwanted features and visualize the data to facilitate data understanding. We propose a new feature selection method using similarity based entropy to overcome the high dimensional data problem. Using 6 datasets with high dimensional feature, we have computed the similarity between feature vector and class vector. Then we find the maximum similarity that can be used for calculating the entropy values of each feature. The selected features are features that having higher entropy than mean entropy of overall features. The fuzzy k-NN classifier was implemented to evaluate the selected features. The experiment result shows that proposed method is able to deal with high dimensional data problem with average accuracy of 80.5%.

Keywords: classification, entropy, feature selection, high dimensional data, similarity

Author: Jayanti Yusmah Sari, Mutmainnah Muchtar, Mohammad Zarkasi, Agus Zainal Arifin

Journal Code: jptkomputergg140016

Artikel Terkait :

Jp Teknik Komputer gg 2014

• Research on Keyhole Diameter’s Vision Measurement Based on Parallel Technology
• Strain Transfer and Test Research of Stick-up Fiber Bragg Grating Sensors
• Matrix Mask Overlapping and Convolution Eight Directions for Blood Vessel Segmentation on Fundus Retinal Image
• Application Status of the Barkhausen Effect in Nondestructive Testing
• The New Complex-Valued Wavelet Neural Network
• A Review of Current Control Strategy for Single-Phase Grid-Connected Inverters
• Wireless Power Transfer by Using Solar Energy
• One-Time Password Implementation on Lego Mindstorms NXT
• Simple HAWT Prototype Efficiency at Small Scale Wind Speed
• Combine Target Extraction and Enhancement Methods to Fuse Infrared and LLL Images
• Batik Image Retrieval Based on Color Difference Histogram and Gray Level Co-Occurrence Matrix
• A Novel Part-of-Speech Set Developing Method for Statistical Machine Translation
• A Different Single-Phase Hybrid Five-Level Voltage-Source Inverter Using DC-Voltage Modules
• Multi Population Evolutionary Programming Approach for Distributed Generation Installation
• Failure Mechanism Analysis and Failure Number Prediction of Wind Turbine Blades
• Searching and Visualization of References in Research Documents
• Self-learning PID Control for X-Y NC Position Table with Uncertainty Base on Neural Network
• Adaptive Control of Space Robot Manipulators with Task Space Base on Neural Network
• Planar Finger-Shaped Antenna Used in Ultra-Wideband Wireless Systems
• Information Interchange Layer based on Classification of Information Use (IU)
• A Novel Multi-focus Image Fusion Method Based on Non-negative Matrix Factorization
• A Decision Tree Based on Spatial Relationships for Predicting Hotspots in Peatlands
• Traveling-Wave-Based Fault Location in Electrical Distribution Systems with Digital Simulations
• Multi-focus Image Fusion with Sparse Feature Based Pulse Coupled Neural Network
• Measuring Information Security Awareness of Indonesian Smartphone Users