Classification with partial labels

In this paper, we address the problem of learning when some cases are fully labeled while other cases are only partially labeled, in the form of partial labels. Partial labels are represented as a set of possible labels for each training example, one of which is the correct label. We introduce a discriminative learning approach that incorporates partial label information into the conventional margin-based learning framework. The partial label learning problem is formulated as a convex quadratic optimization minimizing the L2-norm regularized empirical risk using hinge loss. We also present an efficient algorithm for classification in the presence of partial labels. Experiments with different data sets show that partial label information improves the performance of classification when there is traditional fully-labeled data, and also yields reasonable performance in the absence of any fully labeled data. Categories and Subject Descriptors I.5 [Pattern Recognition]: Design Methodology G...

• Type: Conference
• Year: 2008
• Where: KDD
• Authors: Nam Nguyen, Rich Caruana
• Keywords: Data Mining, KDD 2008, Keywords Partial Labels, Label Learning Problem, Partial Label Information