Want to create interactive content? It’s easy in Genially!

Get started free

MBC-6-M2-R3

CEV PUCE

Created on May 1, 2021

Start designing with a free template

Discover more than 1500 professional designs like these:

Audio tutorial

Pechakucha Presentation

Desktop Workspace

Decades Presentation

Psychology Presentation

Medical Dna Presentation

Geometric Project Presentation

Explore all templates

Transcript

Algoritmos de la Biología Computacional

Evaluación de PSSM.

Evaluación de PSSM

Performance metrics

Regression

  • MSPE
  • MSAE
  • R Square
  • Adjust R Square

Classification

  • precision recall
  • ROC- AUC
  • Accuracy
  • Log- loss

Para evaluar una PSSM debemos medir su desempeño. Debemos medir que tan cercanas están las predicciones de las observaciones. Es decir, que tan cercanos están los valores predichos de los valores observados o esperados. Para ello, usaremos medidas como accuracy, PCC, MSE, MCC, ROC, AUC, entre otras.

Others

  • CV Error
  • Heuristic methods to find K
  • Bleu Score (NPL)

Unsupervised

  • Models
  • Rand Index
  • Mutual
Information

Performance metrics

Medición Predicción0.4050 0.8344 0.9373 1.0000 0.8161 0.6388 0.6752 0.9841 0.0253 0.0000 0.3196 0.5388 0.6764 0.6247 0.1872 0.1921 0.4220 0.6546 0.6545 0.6546 0.7917 0.1342 0.4405 0.3551 0.1548 0.0000 0.2740 0.1993 0.4399 0.6461 0.1725 0.3916 0.0539 0.0000 0.3795 0.5623 0.2242 0.1968 0.3108 0.2114

Coeficiente de correlación de Pearson

Coeficiente de correlación de Pearson

Correlation Coefficient: Simple Definition, Formula, Easy Calculation Steps. (n.d.). Statistics How To. Retrieved April 1, 2021, from https://www.statisticshowto.com/probability-and-statistics/correlation-coefficient-formula/

Error cuadrático medio o mean squared error

Exactitud o accuracy

Ordenar

Correlación de Matthews

False Negative True Negative

True Positive False Negative

Correlación de Matthews

5 8

9 5

ROC curves

False Negative True Negative

True Positive False Negative

ROC curves

False Negative True Negative

True Positive False Negative

ROC curves

AUC = 0.5 AUC = 1.0

ROC curves

Serengil, S. (2020, December 10). A Gentle Introduction to ROC Curve and AUC in Machine Learning. Sefik Ilkin Serengil. https://sefiks.com/2020/12/10/a-gentle-introduction-to-roc-curve-and-auc/

AUC (area under the ROC curve)

En resumen:

MCC y PCC Random = 0.0 Perfecto = 1.0 (or -1) ROC (AUC) Random = 0.5 Perfecto = 1 (or 0)

Referencias

  • Morten Nielsen. (2019, July). Performance measures. 22125 - Algorithms in Bioinformatics, DTU. https://kurser.dtu.dk/course/22125
  • Serengil, S. (2020, December 10). A Gentle Introduction to ROC Curve and AUC in Machine Learning. Sefik Ilkin Serengil. https://sefiks.com/2020/12/10/a-gentle-introduction-to-roc-curve-and-auc/