A Survey of Dimension Reduction Techniques

Fodor, Imola K.

doi:10.2172/15002155

Cited by 787 publications

(500 citation statements)

References 33 publications

(5 reference statements)

Supporting

Mentioning

465

Contrasting

Unclassified

Order By: Relevance

“…Dimension reduction methods are traditionally divided into two groups: feature selection-and feature extraction approaches. Feature selection aims at finding a subset of the measured variables while feature extraction is applying a projection of the multidimensional problem space into a space of fewer dimensions thus resulting in aggregate measures that did not exist in the measured environment [16,17].…”

Section: Experimental Methodsologymentioning

confidence: 99%

Sensor Selection for IT Infrastructure Monitoring

Paljak

Kocsis

Egel

et al. 2010

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

Abstract. Supervisory control is the main means to assure a high level performance and availability of large IT infrastructures. Applied control theory is used in physical and virtualization based clustering, autonomic-, self-healing and cloud computing, but similar problems arise in any distributed environment. The selection of a compact, but sufficiently characteristic set of control variables is one of the core problems both for design and run-time complexity. Most results in the literature are based on a single algorithm for variable selection, but our measurements indicate that no single algorithm can generate faithful estimates for all the different operational domains. We propose to use a combination of different model extraction techniques on benchmark-like data logs. The main advantages of this multi-paradigm approach are twofold: it provides good parameter estimators for predictive control in a simple way; and supports the identification of the actual operational domain facilitating context-aware adaptive control, diagnostics and repair.

show abstract

Section: Experimental Methodsologymentioning

confidence: 99%

Sensor Selection for IT Infrastructure Monitoring

Paljak

Kocsis

Egel

et al. 2010

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

show abstract

“…A variety of statistical techniques have been brought to bear on consolidating or clustering terms [24]. These offer the means to go well beyond consolidation of term variants, drawing upon semantic or syntactic associations.…”

Section: Term Clumpingmentioning

confidence: 99%

“Term clumping” for technical intelligence: A case study on dye-sensitized solar cells

Zhang¹,

Porter²,

Hu³

et al. 2014

Technological Forecasting and Social Change

142

View full text Add to dashboard Cite

Tech Mining seeks to extract intelligence from Science, Technology & Innovation information record sets on a subject of interest. A key set of Tech Mining interests concerns which R&D activities are addressed in the publication and patent abstract records under study. This paper presents six "term clumping" steps that can clean and consolidate topical content in such text sources. It examines how each step changes the content, potentially to facilitate extraction of usable intelligence as the end goal. We illustrate for an emerging technology, dye-sensitized solar cells. In this case we were able to reduce some 90,980 terms & phrases to more user-friendly sets through the clumping steps as one indicator of success. The resulting phrases are better suited to contributing usable technical intelligence than the original results. We engaged seven persons knowledgeable about dye-sensitized solar cells (DSSCs) to assess the resulting content. These empirical results advanced the development of a semi-automated term clumping process that can enable extraction of topical content intelligence.

show abstract

“…Dimensionality reduction techniques have been widely used in machine learning for the purposes of data visualization, data compression, noise removal, pattern recognition, exploratory analysis, and time series prediction. Depending on the nature of the observations, techniques can be classified [19] into linear methods such as Principal Component Analysis (PCA), Factor Analysis (FA) or Projection Pursuit (PP), and non-linear methods such as Independent Component Analysis (ICA) or non-linear PCA, just to cite a few.…”

Section: Learning Methods For Re-identificationmentioning

confidence: 99%

Using latent features for short-term person re-identification with RGB-D cameras

Oliver

Albiol

et al. 2015

Pattern Anal Applic

View full text Add to dashboard Cite

This paper presents a system for people re-identification in uncontrolled scenarios using RGB-depth cameras. Compared to conventional RGB cameras, the use of depth information greatly simplifies the tasks of segmentation and tracking. In a previous work, we proposed a similar architecture where people were characterized using colorbased descriptors that we named bodyprints. In this work, we propose the use of latent feature models to extract more relevant information from the bodyprint descriptors by reducing their dimensionality. Latent features can also cope with missing data in case of occlusions. Different probabilistic latent feature models, such as Probabilistic Principal Component Analysis and Factor Analysis, are compared in the paper. The main difference between the models is how the observation noise is handled in each case. Re-identification experiments have been conducted in a real store where people behaved naturally. The results show that the use of the latent features significantly improves the re-identification rates compared to state-of-the-art works.

show abstract

A Survey of Dimension Reduction Techniques

Cited by 787 publications

References 33 publications

Sensor Selection for IT Infrastructure Monitoring

Sensor Selection for IT Infrastructure Monitoring

“Term clumping” for technical intelligence: A case study on dye-sensitized solar cells

Using latent features for short-term person re-identification with RGB-D cameras

Contact Info

Product

Resources

About