A Hybrid Technique for creating classification model using Random Committee and Voted Perceptron Classifier
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2017, Vol 5, Issue 6
Abstract
Various researchers were experimenting with various manual or automated data sets, sometimes we found that due to high error rate various classification tasks of data mining lands into failure zone. This paper tries to give an hybrid approach for reducing error rate during classification tasks with the help of combination of classifiers. All these algorithms are experimented in weka tool. Each algorithm is performed on same data set. It is illustrated that combination of classifiers work better than individual classifiers on some performance measures.
Authors and Affiliations
Prashant Sahu, Rohit Miri
Keywords
Related Articles
- EP ID EP24398
- DOI -
- Views 274
- Downloads 11
Location Based Service without Reveling Self Privacy Data
In this paper, I proposed to solve problems associated with the location data. The user does not want to send his location data (GPS coordinate) to the server directly, since doing so the server can find the user’s loca...
slugDesign and Optimization of Crankpin - A Review
The performance of any automobile largely depends on its size and working in dynamic conditions. The design of the crankpin considers the dynamic loading and the optimization can lead to a pin diameter satisfying the...
slugEconomic load dispatch problem using Heuristic gravitational search algorithm
This paper describes gravitational search algorithm for solving the non convex Economic load Dispatch (ELD) problem. The main objective of economic load dispatch problem is to generate the required amount of power so th...
Survey on JIT Java Compiler
This paper I present a survey on JIT Java Compiler. This paper provides an overview of JIT Java compiler and optimization of code. This compiler translates bytecode on demand into native code.
Grid Computing: An Introduction
Grid computing is a computing framework to meet growing demands for running heterogeneous grid enables application or the traditional “supercomputer” model for research and industry enterprises requiring increasingly la...