دانلود رایگان مقاله دسته های خوشه بندی در ماشین های بردار پشتیبانی

عنوان فارسی
دسته های خوشه بندی در ماشین های بردار پشتیبانی
عنوان انگلیسی
Clustering categories in support vector machines
صفحات مقاله فارسی
0
صفحات مقاله انگلیسی
10
سال انتشار
2017
نشریه
الزویر - Elsevier
فرمت مقاله انگلیسی
PDF
کد محصول
E4431
رشته های مرتبط با این مقاله
مهندسی کامپیوتر
گرایش های مرتبط با این مقاله
هوش مصنوعی
مجله
مجله امگا - Omega
دانشگاه
اداره آمار و تحقیقات اپراتیو، دانشگاه اسپانیا
کلمات کلیدی
دستگاه بردار پشتیبانی، ویژگی های طبقه بندی شده، طبقه بندی اسپارتی، خوشه بندی، برنامه نویسی محدود شده، برنامه نویسی 0-1
چکیده

Abstract


The support vector machine (SVM) is a state-of-the-art method in supervised classification. In this paper the Cluster Support Vector Machine (CLSVM) methodology is proposed with the aim to increase the sparsity of the SVM classifier in the presence of categorical features, leading to a gain in interpretability. The CLSVM methodology clusters categories and builds the SVM classifier in the clustered feature space. Four strategies for building the CLSVM classifier are presented based on solving: the SVM formulation in the original feature space, a quadratically constrained quadratic programming formulation, and a mixed integer quadratic programming formulation as well as its continuous relaxation. The computational study illustrates the performance of the CLSVM classifier using two clusters. In the tested datasets our methodology achieves comparable accuracy to that of the SVM in the original feature space, with a dramatic increase in sparsity.

نتیجه گیری

5. Conclusions


In this paper the CLSVM methodology is proposed, based on the SVM with the linear kernel and performing a clustering for categorical features and building an SVM classifier in the clustered feature space. Four strategies are presented to build the CLSVM classifier by means of QCQP, MIQP and QP formulations. When using two clusters, the CLSVM classifier has a comparable classi- fication accuracy to the SVMO classifier, in nine of the twelve benchmark datasets. In the remaining three datasets, the CLSVM classifier outperforms the SVMO classifier in two datasets, and is outperformed in the other one. In terms of sparsity of the classifier with respect to the categorical features, the CLSVM methodology shows a dramatic improvement over the SVMO.


بدون دیدگاه