Relevant Label Identification for Multi-Label Image Classification

  Radhika V. Devkar,  Sankirti Shiravale

Area of an image multi-label classification is increase rapidly in recent years, in machine learning and computer vision. Multi-label classification has attracted significant attention from researchers and has been applied to an image annotation. In multi-label classification, each instance is assigned to multiple labels, is a common problem in data analysis. In image classification, images are classified into multi-label classes using multi-label classification approach. However most of the multi-label classification approaches are incapable to explore the inter-label correlation between multiple labels effectively. Inter-label correlation analysis is important in multi-label classification. Using inter-label correlation, images can be classified into multiple classes very effectively. To address inter-label correlation problem in multi-label classification, label graph learning with support vector machine technique is used.

Keywords- Multi-label classification, image annotation, inter-label correlation, support vector machine.

Unique Identification Number - IJEDR1703049
Page Number(s) - 340-346
Pubished in - Volume 5 | Issue 3 | July 2017
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Radhika V. Devkar,  Sankirti Shiravale,   "Relevant Label Identification for Multi-Label Image Classification", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.5, Issue 3, pp.340-346, July 2017, Available at :http://www.ijedr.org/papers/IJEDR1703049.pdf