Download Machine Learning: Supervised & Unsupervised Learning, Classification & Regression and more Essays (high school) Mathematics in PDF only on Docsity!
Assignment
Part
Q1: What do you understand by the Machine Learning?
Machine learning is an application of artificial intelligence (AI) that provides systems the ability to
automatically learn and improve from experience.
Q2: What are the differences between supervised and unsupervised machine learning?
Parameters Supervised machine learning
technique
Unsupervised machine learning technique
Process input and output variables are given. only input data is given
Input Data Algorithms are trained using labeled
data.
Algorithms are used data which is not labeled
Number of
Classes
Number of classes is known. Number of classes is not known.
Q3 : Differentiate between classification and regression in Machine Learning.
PARAMENTER CLASSIFICATION REGRESSION
Basic
mapping values to predefined
classes.
mapping values to continuous
output.
prediction of Discrete values Continuous values
Method of
calculation by measuring accuracy
by measurement of root mean
square error
Q4 : What is Gradient descent means?
Gradient Descent is an optimization algorithm for finding a local minimum of a differentiable
function.
Q5 : Do gradient descent methods always converge to same point?
No, they always don't.
Q6 : What is Overfitting? Why overfitting can be happened? How to be avoided?
Overfitting is a major problem in machine learning.
It happens when a model captures noise instead of signal.
It can be avoided by some methods:
1 Cross-validation:
2 Dimension reduction
3 Regularization
Q7: What is Underfitting? Why Underfitting can be happened? How to be avoided?
Underfitting is a statistical model
when it cannot capture the underlying trend of the data.
Techniques to avoid underfitting:
- Increase model complexity
- Increase number of features, performing feature engineering
- Remove noise from the data.
Q8 : What is data normalization?
Data normalization is the organization of data to appear similar across all records and fields.
Q9 : What are the important outcomes of DT?
1 Focus on the Future
2 Design Thinking
3 Starting Young
Q10 : What the regularization is and why it is useful?
techniques to reduce the error by fitting a function appropriately on the given training set and avoid
overfitting.
Compute Gini for sub-nodes with the formula: The sum of the square of probability for success and
failure (p^2 + q^2)
Compute Gini for split by weighted Gini rate of every node of the split
Now, Entropy is the degree of indecency that is given by the following:
where a and b are the probabilities of success and failure of the node
When Entropy = 0, the node is homogenous
When Entropy is high, both groups are present at 50–50 percent in the node.
Finally, to determine the suitability of the node as a root node, the entropy should be very low.
Q16 : What are the three most important components of every machine learning algorithm?
1 Data
2 Features
3 Algorithms
Q17 : Describe the benefits of regularization?
Can Make Models More Useful by Reducing Overfitting.
Q18 : What is the decision boundary given a logistic function?
The line or margin that separates the classes
Q19 : What are the advantages of data normalization?
The benefits of normalization include:
1 A logical map
2 Data consistency
3 Connection to other systems
4 Increased security
5 Cost savings
Q
ID OXYCON AGE HEART
RATE
predction
s
b.
Error delta Error delta 2
Error delta
Squared
Error
ID OxyCon Prediction Error
sum
Sum of squared errors => 2,017.
.b
RECIDIVIST = true
. c
>25K
> 25K
25K-50K
25K-50K
25K-50K
< 50K
25K-50K
25K-50K
The threshold Age >26 has the highest information gain, and
consequently, it is the best thershold to use if are splitting the
dataset using the Age feature
