A simultaneous learning framework for clustering and classification

Cai, Weiling; Chen, Songcan; Zhang, Daoqiang

doi:10.1016/j.patcog.2008.11.029

Cited by 28 publications

(24 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, their performance degrades if the underlying input-output map changes over time. -There is relatively little work in incorporating both labeled and unlabeled data while building ensembles, in contrast to the substantial amount of recent interest in semisupervised learning-including semisupervised clustering, semisupervised classification, clustering with constraints, and transductive learning methods-using a single model [Chapelle et al 2006;Zhu and Goldberg 2009;Cai et al 2009;Forestier et al 2010;Chen et al 2009]. …”

Section: Related Workmentioning

confidence: 99%

An Optimization Framework for Combining Ensembles of Classifiers and Clusterers with Applications to Nontransductive Semisupervised Learning and Transfer Learning

Acharya

Hruschka

Ghosh

et al. 2014

ACM Trans. Knowl. Discov. Data

View full text Add to dashboard Cite

An optimization framework for combining ensembles of classifiers and clusterers with applications to nontransductive semisupervised learning and transfer learning. Data, New York, v.9, n.1, p Unsupervised models can provide supplementary soft constraints to help classify new "target" data because similar instances in the target set are more likely to share the same class label. Such models can also help detect possible differences between training and target distributions, which is useful in applications where concept drift may take place, as in transfer learning settings. This article describes a general optimization framework that takes as input class membership estimates from existing classifiers learned on previously encountered "source" (or training) data, as well as a similarity matrix from a cluster ensemble operating solely on the target (or test) data to be classified, and yields a consensus labeling of the target data. More precisely, the application settings considered are nontransductive semisupervised and transfer learning scenarios where the training data are used only to build an ensemble of classifiers and are subsequently discarded before classifying the target data. The framework admits a wide range of loss functions and classification/clustering methods. It exploits properties of Bregman divergences in conjunction with Legendre duality to yield a principled and scalable approach. A variety of experiments show that the proposed framework can yield results substantially superior to those provided by naïvely applying classifiers learned on the original task to the target data. In addition, we show that the proposed approach, even not being conceptually transductive, can provide better results compared to some popular transductive learning techniques. ACM Transactions on Knowledge Discovery from

show abstract

Section: Related Workmentioning

confidence: 99%

An Optimization Framework for Combining Ensembles of Classifiers and Clusterers with Applications to Nontransductive Semisupervised Learning and Transfer Learning

Acharya

Hruschka

Ghosh

et al. 2014

ACM Trans. Knowl. Discov. Data

View full text Add to dashboard Cite

show abstract

“…they adopt a two-step learning paradigm which fails to realize the simultaneous optimization for both criteria. This may limit the strength of both clustering and To obtain the satisfactory clustering and classification result and inspired by our previous work [21], we present a multi-objective simultaneous learning framework (named MSCC) for both clustering and classification learning. In its implement, we first employ the Bayesian theory to bridge the connection between both and make all their objectives only dependent on the same set of the cluster centers as the parameters to be optimized.…”

Section: Hidden Layermentioning

confidence: 99%

“…Such kind of algorithms sequentially optimizes the clustering criterion and the classification criterion, and thus fails to achieve the simultaneous optimality for such two criteria. Recently, we have gone a small step ahead in this research and proposed a simultaneous learning algorithm for clustering and classification (named SCC) [21]. In SCC, the classification criterion and clustering criterion are combined to a single objective function by a trade-off parameter, whose goal is to compromise the classification and the clustering performances, but its value in optimizing the objective is generally hard to be optimally chosen except for an exhaustive search in some range, which is a heavier learning burden.…”

Section: Introductionmentioning

confidence: 99%

A Multiobjective Simultaneous Learning Framework for Clustering and Classification

Cai

Chen

Zhang

2010

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

Traditional pattern recognition involves two tasks: clustering learning and classification learning.Clustering result can enhance the generalization ability of classification learning, while the class information can improve the accuracy of clustering learning. Hence, both learning methods can complement each other. To fuse the advantages of both learning methods together, many existing algorithms have been developed in a sequential fusing way by first optimizing the clustering criterion and then the classification criterion associated with the obtained clustering results. However, such kind of algorithms naturally fails to achieve the simultaneous optimality for two criteria, and thus have to sacrifice either the clustering performance or the classification performance. To overcome that problem, in this paper, we present a multi-objective simultaneous learning framework (named MSCC) for both clustering and classification learning. MSCC utilizes multiple objective functions to formulate the clustering and classification problems, respectively, and more importantly it employs the Bayesian theory to make these functions all only dependent on a set of the same parameters, i.e., clustering centers which play a role of the bridge connecting the clustering and classification learning. By simultaneously optimizing the clustering centers embedded in these functions, not only the effective clustering performance but also the promising classification performance can be simultaneously attained. Furthermore, from the multiple Pareto-optimality solutions obtained in MSCC, we can get an interesting observation that there is complementarity to great extent between clustering and classification learning processes. Empirical results on both synthetic and real data sets demonstrate the effectiveness and potential of MSCC.

show abstract

“…These two tasks are usually treated separately, because both goals are quite different. However, when label information is available, simultaneous learning clustering and classification tasks will benefit from each other [5].…”

Section: Introductionmentioning

confidence: 99%

“…This paper focuses on simultaneous learning clustering and classification tasks, hence cluster assumption is used as data prior knowledge. It suggests that when data intrinsic structure revealed by some clustering algorithm is incorporated into the classification task, better performance may be desired [5,13,25].…”

Section: Introductionmentioning

confidence: 99%

Simultaneous clustering and classification over cluster structure representation

Qian

Chen

Cai

2012

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

A simultaneous learning framework for clustering and classification

Cited by 28 publications

References 25 publications

An Optimization Framework for Combining Ensembles of Classifiers and Clusterers with Applications to Nontransductive Semisupervised Learning and Transfer Learning

An Optimization Framework for Combining Ensembles of Classifiers and Clusterers with Applications to Nontransductive Semisupervised Learning and Transfer Learning

A Multiobjective Simultaneous Learning Framework for Clustering and Classification

Simultaneous clustering and classification over cluster structure representation

Contact Info

Product

Resources

About