Abstract
Improving training sets is an area of active research within l to Artificial Intelligence. In particular, it is of particular interest in supervised classification systems, where the quality of training data is crucial. This paper presents a new method for the improvement of training sets, based on approximate sets and artificial ant colonies. The experimental study carried out with international databases allows us to guarantee the quality of the new algorithm, which has a high efficiency.
Highlights
The k Nearest Neighbor classifier is of the peak popular supervised algorithm
The modeling of the graph is given with the following structure, there is a training matrix, which gives way to sub-matrices calculated with typical testers and applying a method of object selection
A new submatrix fusion scheme based on Artificial Ant Colonies was introduced
Summary
The k Nearest Neighbor classifier is of the peak popular supervised algorithm. Nearest Neighbor, to assign the class to a new pattern, compares it. with all the patterns of the training matrix, using a distance function, and determines the k objects of the training matrix closest to the given object. With all the patterns of the training matrix, using a distance function, and determines the k objects of the training matrix closest to the given object. One of the strategies that has been proposed to solve the aforementioned problems of the NN family is the combined or simultaneous selection of features and objects. For this purpose, various methods have been developed [23-27], among which family outstand the fusion submatrices. The latter, despite their multiple advantages and high performance, have not been designed to handle noisy or mislabeled objects, and incorporate very elementary fusion schemes, where the capabilities of the sub-matrices
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Engineering and Advanced Technology
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
