Perform Naive-Bayes classification(fitcnb) with non-zero off-diagonal covariance matrix
5 vues (au cours des 30 derniers jours)
Afficher commentaires plus anciens
simplified
le 18 Jan 2018
Réponse apportée : Ilya
le 19 Jan 2018
Greetings,
I use a Bayesian classification model to generate class-conditional probability density functions (PDFs) from a Monte Carlo (MC) simulation (see Fig 1). The different classes have inter-variable correlations such that the covariance matrix has non-zeros on the off-diagonal elements. However, the Bayesian classification model seems to assume that the off-diagonal elements are zero, such that the PDFs for each class are not shaped according to the MC simulated data (see Fig 2); this makes the PDFs look like ellipsoids that are horizontally aligned.
So, how can I specify the covariance elements in the Bayesian classification model when I for instance want to use it to predict a new data set?
Thanks,
Kenneth
Fig 1:
Fig2:
0 commentaires
Réponse acceptée
the cyclist
le 18 Jan 2018
Modifié(e) : the cyclist
le 18 Jan 2018
Disclaimer: I am not an expert on these methods.
Doesn't the "naive" in naive Bayes specifically mean that the model features are independent from each other (i.e. uncorrelated)? You might need a more sophisticated model.
Plus de réponses (1)
Ilya
le 19 Jan 2018
To estimate covariance per class, use fitcdiscr with discriminant type 'quadratic'.
0 commentaires
Voir également
Catégories
En savoir plus sur Naive Bayes dans Help Center et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!