Understanding Mahalanobis distance criterion for feature selection

Masnan, Maz Jamilah; Mahat, Nor Idayu; Shakaff, Ali Yeon Md; Abdullah, Azizi; Zakaria, N. Z. I.; Yusuf, N.; Subari, Norazian; Zakaria, Ammar; Aziz, Abdul

doi:10.1063/1.4915708

Cited by 11 publications

(11 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Elliptic Envelope belongs to a set of methods with an underlying assumption of known distribution (usually Gaussian) for normal data and all the points distant from the center of the ellipse are considered outliers. The Mahalanobis distance [ 19 ] is used as a measure of distance and an indicator that a given data point may be considered as an outlier.…”

Section: Introductionmentioning

confidence: 99%

Protection of Superconducting Industrial Machinery Using RNN-Based Anomaly Detection for Implementation in Smart Sensor

Wielgosz

Skoczeń

Matteis

2018

Sensors

View full text Add to dashboard Cite

Sensing the voltage developed over a superconducting object is very important in order to make superconducting installation safe. An increase in the resistive part of this voltage (quench) can lead to significant deterioration or even to the destruction of the superconducting device. Therefore, detection of anomalies in time series of this voltage is mandatory for reliable operation of superconducting machines. The largest superconducting installation in the world is the main subsystem of the Large Hadron Collider (LHC) accelerator. Therefore a protection system was built around superconducting magnets. Currently, the solutions used in protection equipment at the LHC are based on a set of hand-crafted custom rules. They were proved to work effectively in a range of applications such as quench detection. However, these approaches lack scalability and require laborious manual adjustment of working parameters. The presented work explores the possibility of using the embedded Recurrent Neural Network as a part of a protection device. Such an approach can scale with the number of devices and signals in the system, and potentially can be automatically configured to given superconducting magnet working conditions and available data. In the course of the experiments, it was shown that the model using Gated Recurrent Units (GRU) comprising of two layers with 64 and 32 cells achieves 0.93 accuracy for anomaly/non-anomaly classification, when employing custom data compression scheme. Furthermore, the compression of proposed module was tested, and showed that the memory footprint can be reduced four times with almost no performance loss, making it suitable for hardware implementation.

show abstract

Section: Introductionmentioning

confidence: 99%

Protection of Superconducting Industrial Machinery Using RNN-Based Anomaly Detection for Implementation in Smart Sensor

Wielgosz

Skoczeń

Matteis

2018

Sensors

View full text Add to dashboard Cite

show abstract

“…After representing acoustic features of pairs of IDS and ADS stimuli by calculating their MFCCs, the acoustic distinctiveness between each pair was measured by calculating MD on the corresponding MFCC matrices (i.e and ) using a 12-dimensional feature space (i.e., 12 MFCCs) (Maesschalck & Massart, 2000; Masnan et al, 2015; Xiang et al, 2008). MD is a multivariate statistical approach that evaluates distances between two multidimensional feature vectors or matrices that belong to two classes (here IDS vs. ADS) (Arjmandi et al, 2018; Heijden, Ferdinand, Ridder, & Tax, 2005; Maesschalck & Massart, 2000; Masnan et al, 2015). The MD calculation returns the distance between means of two classes (here IDS and ADS) relative to the average per-class covariance matrix (Maesschalck & Massart, 2000).…”

Section: Methodsmentioning

confidence: 99%

“…Second, to quantify the acoustic distinctiveness between IDS and ADS, we calculated a Mahalanobis distance (MD) measure over MFCCs features. MD is a multivariate distance metric that has been widely used to measure the distances between vectors in a variety of multidimensional feature spaces (Arjmandi, Dilley, & Wagner, 2018;Masnan et al, 2015;Xiang, Nie, & Zhang, 2008). Third, to understand how the vocoding process may influence intelligibility of caregivers' speech, we calculated the speech-to-reverberation-modulation energy ratio (SRMR) to model the signal-based intelligibility of signals delivered to listeners with NH, as well as its CItailored version (SRMR-CI) to model intelligibility of signals delivered to listeners with CIs (Santos et al, 2013).…”

Section: G Current Studymentioning

confidence: 99%

Section: Acoustic Distinctiveness Quantification Using Mahalanobismentioning

confidence: 99%

“…MD is a multivariate statistical approach that evaluates distances between two multidimensional feature vectors or matrices that belong to two classes (here IDS vs. ADS) (Arjmandi et al, 2018;Heijden, Ferdinand, Ridder, & Tax, 2005;Maesschalck & Massart, 2000;Masnan et al, 2015).…”

Section: Acoustic Distinctiveness Quantification Using Mahalanobismentioning

confidence: 99%

See 2 more Smart Citations

Cochlear implant-related speech processing may diminish the advantage of exposure to infant-directed speech

Arjmandi

Houston

Wang

et al. 2020

Preprint

View full text Add to dashboard Cite

Caregivers modify their speech when talking to infants, a specific type of speech known as infant-directed speech (IDS). This speaking style facilitates language learning compared to adult-directed speech (ADS) in infants with normal hearing (NH). While infants with NH and those with cochlear implants (CIs) prefer listening to IDS over ADS, it is yet unknown how CI speech processing may affect the acoustic distinctiveness between ADS and IDS, as well as the degree of intelligibility of these. This study analyzed speech of seven female adult talkers to investigate the effects of simulated CI speech processing on (1) acoustic distinctiveness between ADS and IDS, (2) estimates of intelligibility of caregivers speech in ADS and IDS, and (3) individual differences in caregivers ADS-to-IDS modification and speech intelligibility. Results suggest that CI speech processing is substantially detrimental to the acoustic distinctiveness between ADS and IDS, as well as to the intelligibility benefit derived from ADS-to-IDS modifications. Moreover, the observed considerable variability across individual talkers in acoustic implementation of ADS-to-IDS modification and speech intelligibility was significantly reduced due to CI speech processing. The findings are discussed in the context of the link between IDS and language learning in infants with CIs.

show abstract

Moral inferences from androgynous faces are beyond categorical uncertainty: Evidence of a positive bias towards androgynous targets

Ansani,

Olivera‐La Rosa

2024

Euro J Social Psych

View full text Add to dashboard Cite

Postulating a negative bias towards social ambiguity, we conducted cross‐cultural online research to assess whether categorical discrepancies in the perception of androgynous faces were associated with the uncanny feeling and inferences of different morality. Across four studies, we found that androgynous faces were harder to classify into a binary sex category than sex‐typical faces, but this difficulty did not influence social judgements of androgynous targets in a negative fashion. In Study 1 (Spanish‐speaking sample, N = 76), we found that androgynous faces were rated as more trustworthy, less creepy, and less morally different than sex‐typical faces. Study 2 replicated most of the findings from Study 1 in an Italian sample (N = 45). Positive bias towards androgyny was not replicated with a different set of stimuli featuring faces of diverse ethnic backgrounds (Study 3, Spanish‐speaking sample, N = 140). However, results revealed a main effect of ethnicity in participants’ responses. When controlling for the effect of morphing procedures in stimuli selection, an overall positive bias towards androgynous targets arose, especially when compared to masculine targets (Study 4, Spanish‐speaking sample, N = 85). These findings suggest that, at least in certain conditions, a positive social bias towards androgynous faces may emerge that does not depend on categorical uncertainty and facial attractiveness.

show abstract

Understanding Mahalanobis distance criterion for feature selection

Cited by 11 publications

References 2 publications

Protection of Superconducting Industrial Machinery Using RNN-Based Anomaly Detection for Implementation in Smart Sensor

Protection of Superconducting Industrial Machinery Using RNN-Based Anomaly Detection for Implementation in Smart Sensor

Cochlear implant-related speech processing may diminish the advantage of exposure to infant-directed speech

Moral inferences from androgynous faces are beyond categorical uncertainty: Evidence of a positive bias towards androgynous targets

Contact Info

Product

Resources

About