Multi level incremental influence measure based classification of medical data for improved classification

The problem of classification in medical data has been well studied and there exist number of solution for the classification of medical using different measures and methods. However, the methods still suffer to achieve higher performance in classification accuracy. Towards the development of classi...

Full description

Saved in:
Bibliographic Details
Published in:Cluster computing 2019-11, Vol.22 (Suppl 6), p.15073-15080
Main Authors: Ananthajothi, K., Subramaniam, M.
Format: Article
Language:English
Subjects:
Accuracy
Algorithms
Cancer
Cardiovascular disease
Classification
Computer Communication Networks
Computer Science
Data mining
Data points
Datasets
Decision trees
Diabetes
Estimates
Heart
Iterative methods
Methods
Neural networks
Operating Systems
Processor Architectures
Statistical analysis
Support vector machines
Typhoid
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The problem of classification in medical data has been well studied and there exist number of solution for the classification of medical using different measures and methods. However, the methods still suffer to achieve higher performance in classification accuracy. Towards the development of classification performance, the Multi Level Incremental Influence Measure (MLIIM) based classification algorithm is presented in this paper. The method preprocess the input data set to fix the noise issue by removing the incomplete data. In the second stage, the method estimates the influence measure in multiple levels at iterative manner. Finally, the method estimates, class influence weight (CIW) for different classes. Based on the computed class influence weight, an target class is selected to assign label to the data point. The proposed algorithm produces efficient classification and increases the classification accuracy.
ISSN:1386-7857
1573-7543
DOI:10.1007/s10586-018-2498-z