ACO Based Feature Subset Selection for Multiple k-Nearest Neighbor Classifiers
Journal Title: International Journal on Computer Science and Engineering - Year 2011, Vol 3, Issue 5
Abstract
The k-nearest neighbor (k-NN) is one of the most popular algorithms used for classification in various fields of pattern recognition & data mining problems. In k-nearest neighbor classification, the result of a new instance query is classified based on the majority of k-nearest neighbors. Recently researchers have begun paying attention to combining a set of individual k-NN classifiers, each using a different subset of features, with the hope of improving the overall classification accuracy. In this paper we proposed Ant Colony Optimization (ACO) based feature subset selection for multiple k-nearest neighbor classifiers. The ACO is an iterative meta-heuristic search technique, which inspired by the foraging food behavior of real ant colonies. In ACO, real ants become artificial ants with the particular abilities such as distance determination & tour memory. The solution is constructed in a probabilistic way based on pheromone model in the form of numerical values. The concept of this approach is selecting the best possible subsets of feature from the original set with the help of ACO and combines the outputs from multiple k-NN classifiers. The experimental results show that this proposed method improves the average classification accuracy of k-NN classifier.
Authors and Affiliations
Shailendra Kumar Shrivastava , Pradeep Mewada
Keywords
Related Articles
- EP ID EP160557
- DOI -
- Views 135
- Downloads 0
A Service Oriented Architecture to Integrate Short Message Service (SMS) Notification in Road Traffic Volume Control System
The traffic volume becomes one of the top problems in the world, the volume of traffic spends many time and much money, the traffic volume grows daily, and there is not effective and suitable solutions for grows problem....
Customer Relationship Management System with USCM-AKD Approach of D3M
Customer Relationship Management (CRM for short) System emerged in the last decade to reflect the central role of the customer for the strategic positioning of a company. One of the most significant changes in the practi...
Generation of a pool of variable size symmetric keys through Image
This paper introduces a new concept of the generation of a unending pool of keys through an image leaving behind the idea of sending keys every time for encryption and decryption. This can help in avoiding the problem of...
“Economic Load Dispatch by Generating Units under Varying Load Demands Using Artificial Neural Network”
The problem of economic load dispatch has been tested on a generating plant having Ten units of generators operating in parallel. What shall be the load allocations to individual units for given load demand on the plant?...
Design and Simulation for Producing Two Amplitude Matched Anti-phase Sine Waveforms Using ±2.5 V CMOS Current-Mode Approach
In this paper the current mode approach called “Current Conveyor (CCII+)” has been incorporated to design and simulate the circuit for producing two amplitude matched anti-phase sine waveforms which are frequently used i...