A Comparison on Score Spaces for Expression Microarray Data Classification

Abstract-A score function induced by a generative model of the data can provide a feature vector of a fixed dimension for each data sample. Data samples themselves may be of differing lengths (e.g., speech segments or other sequential data), but as a score function is based on the properties of the data generation process, it produces a fixed-length vector in a highly informative space, typically referred to as "score space." Discriminative classifiers have been shown to achieve higher performances in appropriately chosen score spaces with respect to what is achievable by either the corresponding generative likelihood-based classifiers or the discriminative classifiers using standard feature extractors. In this paper, we present a novel score space that exploits the free energy associated with a generative model. The resulting free energy score space (FESS) takes into account the latent structure of the data at various levels and can be shown to lead to classification performance that at least matches the performance of the free energy classifier based on the same generative model and the same factorization of the posterior. We also show that in several typical computer vision and computational biology applications the classifiers optimized in FESS outperform the corresponding pure generative approaches, as well as a number of previous approaches combining discriminating and generative models.

show abstract

“…An experimental comparison on score spaces extracted from topic models in the context of microarray data classification can be found in [28].…”

Section: Staged Methods and Score Spacesmentioning

confidence: 99%

Free Energy Score Spaces: Using Generative Information in Discriminative Classifiers

Perina

Cristani

Castellani

et al. 2012

IEEE Trans. Pattern Anal. Mach. Intell.

Self Cite

View full text Add to dashboard Cite

show abstract

“…-Since, as a base level, we are mostly interested in the quality of unsupervised learning of the distributions over the samples, the whole dataset has been used to train a CG (of course labels are ignored in this phase), in a transductive way [13,4]. As explained in the methodological section, here we employed bidimensional squared Counting Grid models (in principle, also higher dimensional/not squared grids can be used, see [1]).…”

Section: Methodsmentioning

confidence: 99%

“…A very popular example is the Bag of Words approach, where objects are represented as disorganized bags of basic components such as the words of a dictionary. This approach has been succesfully employed in very different applicative domains like computer vision for 2D image or 3D shape retrieval, in bioinformatics for microarray classification, or in medical domain for brain disease detection [2][3][4][5][6][7][8]. However, the Bag of Words (BoW) method has some disadvantage since in many situations it looses a lot of important information.…”

Section: Introductionmentioning

confidence: 99%

Exploiting Geometry in Counting Grids

Perina

Bicego

Castellani

et al. 2013

Similarity-Based Pattern Recognition

Self Cite

View full text Add to dashboard Cite

Abstract. In this paper we exploit the use of known information about the geometry structure of a recently proposed generative model, namely Counting Grid (CG) [1] to improve the performance of classification accuracy. Once the generative model is trained, the geometric structure of the model introduces a natural spatial relations among the estimated latent variables. Such relation is generally ignored when standard maximum likelihood approach (or classical hybrid generative-discriminative approach) is employed for classification purpose. In this work, we propose to take into account the geometric relations of the generative model by proposing an ad hoc similarity measure for CG. In particular, the values relative to each point of the grid is spread around its neighborhood by using information coming from the CG training phase. The proposed approach is succesfully applied in two applicative scenarios: expression microarray classification and MRI brain classification. Experiments show a drastic improvement over standard schemes when our approach is employed.

show abstract

“…This choice is often motivated by the difficulty or computational efficiency of modeling the known structure of the data. Concerning microarray, it has been shown in [12][13][14] that the bag-of-features representation is well-suited also for microarray data, providing interpretable and descriptive signatures. Each sample can be seen as an independent observation; the gene expression value is then interpreted as the "count" of that gene in the sample: the higher the expression level, the "more present" the gene is in such experiment.…”

Section: Background: Counting Grid Modelmentioning

confidence: 99%

“…Since, as a base level, we are mostly interested in the quality of unsupervised learning of the distributions over the microarray samples, the whole dataset has been used to train a CG (of course labels are ignored in this phase), in a transductive way [14,17]. Then, in order to have a fair comparison with the state-of-the-art, we adopted the testing protocol of [8]: the data set was randomly split 2:3/1:3 (training/testing).…”

Section: Experimental Evaluationmentioning

confidence: 99%

Feature Selection Using Counting Grids: Application to Microarray Data

Lovato

Bicego

Cristani

et al. 2012

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. In this paper a novel feature selection scheme is proposed, which exploits the potentialities of a recent probabilistic generative model, the Counting Grid. This model is able to cluster together similar observations, highlighting the compactness of a class and its underlying structure. The proposed feature selection scheme is applied to the expression microarray scenario, a peculiar context with very few patterns and a huge number of features. Experiments on benchmark datasets show that the proposed approach is effective and stable, assessing state-of-the-art classification accuracies.

show abstract

A Comparison on Score Spaces for Expression Microarray Data Classification

Cited by 4 publications

References 23 publications

Free Energy Score Spaces: Using Generative Information in Discriminative Classifiers

Free Energy Score Spaces: Using Generative Information in Discriminative Classifiers

Exploiting Geometry in Counting Grids

Feature Selection Using Counting Grids: Application to Microarray Data

Contact Info

Product

Resources

About