首页|Study Results from Faculty of Mathematical Sciences Provide New Insights into Su pport Vector Machines (Noisy label relabeling by nonparallel support vector mach ine)
Study Results from Faculty of Mathematical Sciences Provide New Insights into Su pport Vector Machines (Noisy label relabeling by nonparallel support vector mach ine)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators publish new report on . According to news reporting originating from Rasht, Iran, by NewsRx corresponden ts, research stated, "In machine learning, models are derived from labeled train ing data where labels signify classes and features define sample attributes. How ever, noise from data collection can impair the algorithm's performance." Our news correspondents obtained a quote from the research from Faculty of Mathe matical Sciences: "Blanco, Japon, and Puerto proposed mixed-integer programming (MIP) models within support vector machines (SVM) to handle label noise in train ing datasets. Nonetheless, it is imperative to underscore that their models demo nstrate an observable escalation in the number of variables as sample size incre ases. The nonparallel support vector machine (NPSVM) is a bi-nary classification method that merges the strengths of both SVM and twin SVM. It accomplishes this by determining two nonparallel hyperplanes by solving two optimization problems . Each hyperplane is strategically po-sitioned to be closer to one of the classe s while maximizing its distance from the other class. In this paper, to take adv antage of NPSVM's fea-tures, NPSVM-based relabeling (RENPSVM) MIP models are dev eloped to deal with the label noises in the dataset. The proposed model adjusts observation labels and seeks optimal solutions while minimizing compu-tational c osts by selectively focusing on class-relevant observations within an e-intensiv e tube." According to the news reporters, the research concluded: "Instances exhibiting s imilarities to the other class are excluded from this e-intensive tube. Experime nts on 10 UCI datasets show that the proposed NPSVM-based MIP models outperform their counter-parts in accuracy and learning time on the majority of datasets."
Faculty of Mathematical SciencesRashtIranAsiaEmerging TechnologiesMachine LearningSupport Vector MachinesV ector Machines
NETL
NSTL
