The Fisher’s Linear Discriminant

Iatan, Iuliana F.

doi:10.1007/978-3-642-14746-3_43

Cited by 10 publications

(7 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generally, LDA is the best known and more widely used method to highlight differences between groups and to classify them. Moreover, it is known that LDA is an important parametric method useful to discriminate samples when the sample allocation is just known [41].…”

Section: Resultsmentioning

confidence: 99%

Varietal Authentication of Extra Virgin Olive Oils by Triacylglycerols and Volatiles Analysis

Blasi

Pollini

Cossignani

2019

Foods

View full text Add to dashboard Cite

In recent years, there is an increasing interest in high-quality extra virgin olive oils (EVOOs) produced from local cultivars. They have particular chemical/organoleptic characteristics and are frequently subjected to fraud, whereby the control of quality requires a powerful varietal check. In the present research, triacylglycerols (TAGs) and volatiles have been studied as chemical markers for the authentication of EVOO samples from four Italian varieties of Olea europea (Dolce Agogia, Frantoio, Leccino, and Moraiolo). The monocultivar EVOO samples have been subjected to a chemical–enzymatic chromatographic method in order to perform a stereospecific analysis, an important procedure for the characterization of TAG of food products. The results, combined with chemometric analysis (linear discriminant analysis, LDA), were elaborated in order to classify Italian EVOO monocultivar samples. In accordance with the total and intrapositional fatty acid (FA) composition of TAG fraction, the results were allowed to carry out a varietal discrimination. In addition, volatile compounds were also determined by solid-phase micro-extraction gas chromatography–mass spectrometry analysis. All EVOO samples were correctly classified when TAG stereospecific data and volatile results were elaborated by the LDA procedure, even if volatile compounds showed a higher discriminant power.

show abstract

Section: Resultsmentioning

confidence: 99%

Varietal Authentication of Extra Virgin Olive Oils by Triacylglycerols and Volatiles Analysis

Blasi

Pollini

Cossignani

2019

Foods

View full text Add to dashboard Cite

show abstract

“…The post hoc analyses were performed using linear discriminant analysis (LDA). LDA is a multivariate method, a generalization of Fisher's linear discriminant (Iatan, 2010), used to find linear combinations of DVs that best separate the groups. In our context, LDA delivers a finite set of vectors (termed canonical vectors) with subject corresponding values.…”

Section: Substance Effects In Dfnc Strengthmentioning

confidence: 99%

The Impact of Combinations of Alcohol, Nicotine, and Cannabis on Dynamic Brain Connectivity

Vergara

Weiland

Hutchison

et al. 2017

Neuropsychopharmacol.

View full text Add to dashboard Cite

Alcohol, nicotine, and cannabis are among the most commonly used drugs. A prolonged and combined use of these substances can alter normal brain wiring in different ways depending on the consumed cocktail mixture. Brain connectivity alterations and their change with time can be assessed using functional magnetic resonance imaging (fMRI) because of its spatial and temporal content. Here, we estimated dynamic functional network connectivity (dFNC) as derived from fMRI data to investigate the effects of single or combined use of alcohol, nicotine, and cannabis. Data from 534 samples were grouped according to their substance use combination as controls (CTR), smokers (SMK), drinkers (DRN), smoking-and-drinking subjects (SAD), marijuana users (MAR), smoking-and-marijuana users (SAM), marijuana-and-drinking users (MAD), and users of all three substances (ALL). The DRN group tends to exhibit decreased connectivity mainly in areas of sensorial and motor control, a result supported by the dFNC outcome and the alcohol use disorder identification test. This trend dominated the SAD group and in a weaker manner MAD and ALL. Nicotine consumers were characterized by an increment of connectivity between dorsal striatum and sensorimotor areas. Where possible, common and separate effects were identified and characterized by the analysis of dFNC data. Results also suggest that a combination of cannabis and nicotine have more contrasting effects on the brain than a single use of any of these substances. On the other hand, marijuana and alcohol might follow an additive effect trend. We concluded that all of the substances have an impact on brain connectivity, but the effect differs depending on the dFNC state analyzed.

show abstract

“…We then apply Fisher linear discriminant as our objective function (Iatan 2010). First, we compute the class covariance:…”

Section: Methodsmentioning

confidence: 99%

Neuronal pattern separation of motion-relevant input in LIP activity

Berberian

MacPherson

Giraud

et al. 2017

Journal of Neurophysiology

View full text Add to dashboard Cite

In various regions of the brain, neurons discriminate sensory stimuli by decreasing the similarity between ambiguous input patterns. Here, we examine whether this process of pattern separation may drive the rapid discrimination of visual motion stimuli in the lateral intraparietal area (LIP). Starting with a simple mean-rate population model that captures neuronal activity in LIP, we show that overlapping input patterns can be reformatted dynamically to give rise to separated patterns of neuronal activity. The population model predicts that a key ingredient of pattern separation is the presence of heterogeneity in the response of individual units. Furthermore, the model proposes that pattern separation relies on heterogeneity in the temporal dynamics of neural activity and not merely in the mean firing rates of individual neurons over time. We confirm these predictions in recordings of macaque LIP neurons and show that the accuracy of pattern separation is a strong predictor of behavioral performance. Overall, results propose that LIP relies on neuronal pattern separation to facilitate decision-relevant discrimination of sensory stimuli. A new hypothesis is proposed on the role of the lateral intraparietal (LIP) region of cortex during rapid decision making. This hypothesis suggests that LIP alters the representation of ambiguous inputs to reduce their overlap, thus improving sensory discrimination. A combination of computational modeling, theoretical analysis, and electrophysiological data shows that the pattern separation hypothesis links neural activity to behavior and offers novel predictions on the role of LIP during sensory discrimination.

show abstract

The Fisher’s Linear Discriminant

Cited by 10 publications

References 4 publications

Varietal Authentication of Extra Virgin Olive Oils by Triacylglycerols and Volatiles Analysis

Varietal Authentication of Extra Virgin Olive Oils by Triacylglycerols and Volatiles Analysis

The Impact of Combinations of Alcohol, Nicotine, and Cannabis on Dynamic Brain Connectivity

Neuronal pattern separation of motion-relevant input in LIP activity

Contact Info

Product

Resources

About