Classification of Multi-labeled Data: A Generative Approach

Abstract

Multi-label classification assigns a data item to one or several classes. This problem of multiple labels arises in fields like acoustic and visual scene analysis, news reports and medical diagnosis. In a generative framework, data with multiple labels can be interpreted as additive mixtures of emissions of the individual sources. We propose a deconvolution approach to estimate the individual contributions of each source to a given data item. Similarly, the distributions of multi-label data are computed based on the source distributions. In experiments with synthetic data, the novel approach is compared to existing models and yields more accurate parameter estimates, higher classification accuracy and ameliorated generalization to previously unseen label sets. These improvements are most pronounced on small training data sets. Also on real world acoustic data, the algorithm outperforms other generative models, in particular on small training data sets.

Multi‐label learning: a review of the state of the art and ongoing research

“…It captured the underlying structures via the latent random variables in a supervised manner. In Additive‐Generative Multi‐Label Model ( MAd Gen ), a deconvolution approach estimated the individual contribution of each label to a given data item and, in Ref , a set of three models based on the Latent Dirichlet Allocation (LDA) framework was presented.…”

Performance metrics for the evaluation of hyperspectral chemical identification systems

“…[24][25][26][27][28][29] For a single dataset, the CM has all the performance information available in detail. However, for ease of interpretation and comparing algorithms, the CM is summarized using several scalar performance metrics that are appropriate for detection and identification.…”

Multi-label classification with Bayes' theorem

“…Some researchers assume that all single labels are independent, and multi-labels are selected by selecting labels with high individual confidence values ( [26]). Other researchers utilize a mixture model to combine single label models ( [8], [10], [11], [17], [23], [25]). In [10], McCallum presented a Bayesianbased model for multi-label.…”

“…There are two typical transformation methods, Binary Relevance (BR) and Label Powerset (LP) ( [18], [20]). In BR, labels are assumed to be independent ( [7], [8], [10], [11], [17], [23]). Every label has its own single-label classifier.…”