Data Science Final Exam Questions
with 100% Verified Answers
What are normal (Gaussian) distribution defined by? - ✔ ✔ The mean and
the variance
Maximum Likelihood Principle - ✔ ✔ Used to estimate the parameters of a
stats model by finding the distribution that makes data most likely to appear.
What does it mean to be normally distributed? - ✔ ✔ Most measurements
are close to the mean and relatively symmetrical.
Log Likelihoods - ✔ ✔ Maximizing the likelihood is the same as maximizing the
log likelihood
maximum likelihood Regression - ✔ ✔ Estimating conditions expectation so that
y hat is the estimate of Y|X=x.
, Why is maximum likelihood good? - ✔ ✔ - way of adapting the model to fit
many different types of observations
- good theretical properties
- used as a foundation that can be adjusted
Machine Learning - ✔ ✔ The process of programming an alg to automatically
learn form data.
Why would you want to use a learning algorithm? - ✔ ✔ - Hard to code by hand
- Needs to adapt to environment
- Need to be better than humans
- Privacy & fairness
Input - ✔ ✔ A piece of data
Output - ✔ ✔ A prediction or decision
3 Types of Machine Learnign - ✔ ✔ 1. Supervised
2. Unsupervised
3. Reinforcement
Supervised Learning - ✔ ✔ Learning from labelled examples
with 100% Verified Answers
What are normal (Gaussian) distribution defined by? - ✔ ✔ The mean and
the variance
Maximum Likelihood Principle - ✔ ✔ Used to estimate the parameters of a
stats model by finding the distribution that makes data most likely to appear.
What does it mean to be normally distributed? - ✔ ✔ Most measurements
are close to the mean and relatively symmetrical.
Log Likelihoods - ✔ ✔ Maximizing the likelihood is the same as maximizing the
log likelihood
maximum likelihood Regression - ✔ ✔ Estimating conditions expectation so that
y hat is the estimate of Y|X=x.
, Why is maximum likelihood good? - ✔ ✔ - way of adapting the model to fit
many different types of observations
- good theretical properties
- used as a foundation that can be adjusted
Machine Learning - ✔ ✔ The process of programming an alg to automatically
learn form data.
Why would you want to use a learning algorithm? - ✔ ✔ - Hard to code by hand
- Needs to adapt to environment
- Need to be better than humans
- Privacy & fairness
Input - ✔ ✔ A piece of data
Output - ✔ ✔ A prediction or decision
3 Types of Machine Learnign - ✔ ✔ 1. Supervised
2. Unsupervised
3. Reinforcement
Supervised Learning - ✔ ✔ Learning from labelled examples
