Ungraded quiz Unit 6.

Slides:



Advertisements
Similar presentations
Ensemble Methods An ensemble method constructs a set of base classifiers from the training data Ensemble or Classifier Combination Predict class label.
Advertisements

A Quick Overview By Munir Winkel. What do you know about: 1) decision trees 2) random forests? How could they be used?
Sparse vs. Ensemble Approaches to Supervised Learning
Resampling techniques Why resampling? Jacknife Cross-validation Bootstrap Examples of application of bootstrap.
Ensemble Learning: An Introduction
Prediction Methods Mark J. van der Laan Division of Biostatistics U.C. Berkeley
Ensemble Learning (2), Tree and Forest
Zhangxi Lin ISQS Texas Tech University Note: Most slides are from Decision Tree Modeling by SAS Lecture Notes 6 Ensembles of Trees.
ENSEMBLE LEARNING David Kauchak CS451 – Fall 2013.
Montecarlo Simulation LAB NOV ECON Montecarlo Simulations Monte Carlo simulation is a method of analysis based on artificially recreating.
Bootstrapping (And other statistical trickery). Reminder Of What We Do In Statistics Null Hypothesis Statistical Test Logic – Assume that the “no effect”
LOGO Ensemble Learning Lecturer: Dr. Bo Yuan
Inference for Regression Simple Linear Regression IPS Chapter 10.1 © 2009 W.H. Freeman and Company.
Ensembles. Ensemble Methods l Construct a set of classifiers from training data l Predict class label of previously unseen records by aggregating predictions.
ISQS 6347, Data & Text Mining1 Ensemble Methods. ISQS 6347, Data & Text Mining 2 Ensemble Methods Construct a set of classifiers from the training data.
Konstantina Christakopoulou Liang Zeng Group G21
Random Forests Ujjwol Subedi. Introduction What is Random Tree? ◦ Is a tree constructed randomly from a set of possible trees having K random features.
Classification Ensemble Methods 1
Data Analytics CMIS Short Course part II Day 1 Part 3: Ensembles Sam Buttrey December 2015.
Ensemble Methods Construct a set of classifiers from the training data Predict class label of previously unseen records by aggregating predictions made.
Decision Trees IDHairHeightWeightLotionResult SarahBlondeAverageLightNoSunburn DanaBlondeTallAverageYesnone AlexBrownTallAverageYesNone AnnieBlondeShortAverageNoSunburn.
1 Chapter 8: Model Inference and Averaging Presented by Hui Fang.
Tree and Forest Classification and Regression Tree Bagging of trees Boosting trees Random Forest.
Bootstrapping and Randomization Techniques Q560: Experimental Methods in Cognitive Science Lecture 15.
Statistics 350 Lecture 2. Today Last Day: Section Today: Section 1.6 Homework #1: Chapter 1 Problems (page 33-38): 2, 5, 6, 7, 22, 26, 33, 34,
1 Machine Learning Lecture 8: Ensemble Methods Moshe Koppel Slides adapted from Raymond J. Mooney and others.
Bias-Variance Analysis in Regression  True function is y = f(x) +  where  is normally distributed with zero mean and standard deviation .  Given a.
Decision tree and random forest
Ensemble Classifiers.
Machine Learning: Ensemble Methods
Archival research: Ungraded review questions
Bagging and Random Forests
Confidence Interval Estimation
Zaman Faisal Kyushu Institute of Technology Fukuoka, JAPAN
Eco 6380 Predictive Analytics For Economists Spring 2016
Chapter 13 – Ensembles and Uplift
Ensemble Learning Introduction to Machine Learning and Data Mining, Carla Brodley.
ECE 5424: Introduction to Machine Learning
Ensemble methods: Bagging and boosting
Ungraded quiz Unit 3.
A “Holy Grail” of Machine Learing
Estimates of Bias & The Jackknife
Ungraded quiz Unit 1.
Data Mining Practical Machine Learning Tools and Techniques
Ungraded quiz Unit 1.
Direct or Remotely sensed
Ungraded quiz Unit 5.
Predictive distributions
Methods of Economic Investigation Lecture 12
Introduction to Data Mining, 2nd Edition
Ungraded quiz Unit 4.
Ch13 Empirical Methods.
Siyan Gan, Pepperdine University
Ensembles.
Bootstrapping Jackknifing
Ensemble learning.
Cross-validation Brenda Thomson/ Peter Fox Data Analytics
Ensemble learning Reminder - Bagging of Trees Random Forest
Ensemble forecasts and seasonal precipitation tercile probabilities
Ensemble Methods: Bagging.
Simulation Berlin Chen
Ungraded quiz Unit 6.
Ungraded quiz Unit 7.
Ungraded quiz Unit 3.
Ungraded quiz Unit 1.
Ungraded quiz Unit 1.
Ungraded quiz Unit 2.
Advisor: Dr.vahidipour Zahra salimian Shaghayegh jalali Dec 2017
Ungraded quiz Unit 3.
Ensemble methods: Bagging and boosting
Presentation transcript:

Ungraded quiz Unit 6

Show me your fingers Do not shout out the answer, or your classmates will follow what you said. Use your fingers One finger (the right finger) = A Two fingers = B Three fingers = C Four fingers = D No finger = I don’t know. I didn’t study

Which of the following statements about resampling is true? Originally resampling was developed for big data sets. Two resampling methods, Jackknife and exact tests, are commonly used in big data analytics. The root of resampling is Monte Carlo simulation. Resampling does not need real data.

Which of the following statements is true? Like the sampling distribution, the empirical distribution generated by bootstrapping is always normal. Jackknife does not assume that every observation is treated equally. Bootstrapping means one sample gives rise to many samples. Fisher used the exact to solve the problem of lady’s tasting coffee.

Which of the following is true? When the bias is high, the model is unstable. When the model is simple, the variance is high. Bias is the distance between the actual and the estimated values. Bias results from random noise.

Which of the following is true? Bagging reduces bias. Boosting reduces variance. Ensemble methods cancel out errors by resampling. A high-variance model is an over-fitted model.

Which of the following is true? In JMP bootstrap forest randomly subsets the sample and randomly subsets the predictors. The term “random forest” is trademarked by the inventor and thus no other software can use this name. In IBM SPSS Modeler random forest is called bootstrap tree.

What is the typical portion of the out of bag sample (OOBS)? 25% 30%-33% 50% 75%

What is the difference between bagging and boosting? Bagging is sequential ad boosting is a 2-step process. Bagging uses sampling without replacement whereas boosting uses sampling with replacement. In bagging each model is independent but in boosting the previous model informs the next one. Boosting uses weighted average to create a converged model.

What of the following is true? In bagging and boosting the contribution of the variable is measured by the number of splits and the portion. When the DV is binary, we should use average predicted values. When the DV is continuous, we should use majority voting. When the DV is continuous, we report G2.

Which of the following statement is a criterion for model comparison? Entropy R2 RMSE AUC Misclassification rate All of the above