#
*Machine
learning quiz questions TRUE or FALSE with answers, important machine
learning interview questions for data science, Top 3 machine learning
question set*

##
__Machine Learning TRUE / FALSE Questions - SET 02__

###
*1. In general when are
trying to learn an HMM with a small number of states from a large number of
observations, we can almost always increase the training data likelihood by
permitting more hidden states.*

*1. In general when are trying to learn an HMM with a small number of states from a large number of observations, we can almost always increase the training data likelihood by permitting more hidden states.*

(a) TRUE (b)
FALSE

**View Answer**Answer: TRUETo model any
finite length sequence, we can increase the number of hidden states in an HMM
to be the number of observations in the sequence and therefore (with
appropriate parameter choices) generate the observed sequence with
probability 1. Given a fixed number of finite sequences (say n), we would
still be able to assign probability 1/n for generating each sequence. This is
not useful, of course, but highlights the fact that the complexity of HMMs is
not limited. |

###
*2. Assuming a fixed number of attributes, a
Gaussian-based Bayes optimal classifier can be learned in time linear in the
number of records in the dataset.*

*2. Assuming a fixed number of attributes, a Gaussian-based Bayes optimal classifier can be learned in time linear in the number of records in the dataset.*

(a) TRUE (b)
FALSE

**View Answer**Answer: TRUE |

###
*3. Random forests
usually perform better than AdaBoost when your dataset has mislabeled data
points.*

*3. Random forests usually perform better than AdaBoost when your dataset has mislabeled data points.*

(a) TRUE (b)
FALSE

**View Answer**Answer: TRUERandom forest is
highly accurate and robust against noise and outliers. The main advantage of
random forest is that it is less affected by noise. It tries to reduce
variance. AdaBoost shows
poor performance if the data are noisy.Compared to
random forests, AdaBoost performs worse when irrelevant features are included
in the model. |

**************************

###
**Related links:**

**Related links:**

## No comments:

## Post a comment