Open Access
Subscription or Fee Access
A Classification Model Based on Improved Reject Subspace Density Covering Algorithm
Yuan Zhang, Yan-ping Zhang, Fu-lan Qian, Qing-wei Shen
Abstract
Data set classification is the main function of machine learning, and high dimensional characteristics of sample data sets has been one of the key point of machine learning research. In order to solve this problem, this paper puts forward a classification model based on improved reject subspace density covering. Firstly, it adopts optimized density covering ideas to do subspace classification for data set; Secondly, it continues to use the idea of subspace to classify cover sample points whose density is less than the threshold value until all points are covered effectively; Finally, in view of the reject sample data in algorithm test, overall considering the sample number inside of the subspace, explosion radius and the relationship between explosion radius and reject point to give out a new sample rejection strategy, the algorithm can be divided into two parts, training and testing. Through the experimental results we know that the algorithm can improve the generalization ability of classifier on the performance of high-dimensional data sets and noise data set, it has good accuracy which can provide a new way for high-dimensional or large scale data containing noise classification.
Keywords
subspace, covering algorithm, density optimization, sample rejection classification, generalization capability.
Full Text:
PDF
Disclaimer/Regarding indexing issue:
We have provided the online access of all issues and papers to the indexing agencies (as given on journal web site). It’s depend on indexing agencies when, how and what manner they can index or not. Hence, we like to inform that on the basis of earlier indexing, we can’t predict the today or future indexing policy of third party (i.e. indexing agencies) as they have right to discontinue any journal at any time without prior information to the journal. So, please neither sends any question nor expects any answer from us on the behalf of third party i.e. indexing agencies.Hence, we will not issue any certificate or letter for indexing issue. Our role is just to provide the online access to them. So we do properly this and one can visit indexing agencies website to get the authentic information. Also: DOI is paid service which provided by a third party. We never mentioned that we go for this for our any journal. However, journal have no objection if author go directly for this paid DOI service.