A Support Vector Machine and Information Gain based Classification Framework for Diabetic Retinopathy Images
Journal Title: INTERNATIONAL JOURNAL OF COMPUTER TRENDS & TECHNOLOGY - Year 2014, Vol 8, Issue 2
Abstract
Image mining is the process of applying data analysis and discovery algorithms over large volume of image data. It has especially become popular in the fields of forensic sciences, fraud analysis and health care, for it reduces costs in time and money. It allows for the identification of natural group of patients given inputs such as symptoms and further classifies or predicts derivatives from the given data. It couples traditional manual medical data analysis with data mining methods for efficient computer assisted analysis. In this work, the concept of classifying the medical data with and without feature selection technique is discussed. The features representing the useful information about the images were extracted and fed to the mining process. The experimental results demonstrate that the SVM classifier can effectively and efficiently classify the data when compared to other classification algorithms. The information gain based attribute selection method provides the results similar to SVM classifier.
Authors and Affiliations
M. Dharani , T. Menaka , G. Vinodhini
Keywords
Related Articles
- EP ID EP152371
- DOI -
- Views 107
- Downloads 0
A Study on Educational Data Mining
Educational Data Mining is an emerging discipline, concerned with developing methods for exploring the unique types of data that come from educational settings, and using those methods to better understand students, and...
Affective Embedded Systems: a Requirement Engineering Approach
This paper proposes an embedded system development process able to deal with affectivity requirements. This process includes W3C tools, such as EmotionML and SysML, in order to specify cognition and affectivity as system...
SVM Based Ranking Model for Dynamic Adaption to Business Domains
Huge information can be obtained from vertical search domains. Often the information is very large in such a way that users need to browse further to get the required piece of information. In this context ranking plays a...
Survey on End-To-End Confidentiality in Wireless Sensor Networks
A wireless sensor network (WSN) is a collection of a large number of sensor nodes and few sink nodes that have limited Computation, communication and power resources. Data aggregation is used to reduce amount of data tra...
Design an Algorithm for Data Encryption and Decryption Using Pentaoctagesimal SNS
Encryption is the most effective way of computer science concerned with developing schemes and formula to achieve data and information security through the use of codes. Today the privacy is the main issue to sending inf...