Lecture 1 - Introduction

Question 1

What is machine learning about

Answer 1

Machine learning is about using the right features to build the right models that achieve the right tasks

Question 2

What are Tasks

Answer 2

Problems that can be solved with machine learning

Question 3

What are Predictive Tasks

Answer 3

Predicting a target variable from a number of features

Question 4

What are descriptive tasks

Answer 4

exploiting the underlying structure of the data

Question 5

What are 3 Predictive Tasks

Answer 5

• Classification
• Regression
• Predictive clustering

Question 6

What are 3 Descriptive Tasks

Answer 6

• Descriptive clustering
• Association rule mining
• Subgroup Discovery

Descriptive tasks

Question 7

What is the difference between predictive and descriptive tasks

Answer 7

Model output of predictive models involves a target variable, while the model output of the descriptive models does not.

Question 8

Predictive Classification

What is Classification

Also give 2 examples

Answer 8

Classification tasks predict categorical target variable from a set of features.
* image classification
* weather type prediction

Question 9

Predictive regression

What is regression

Give 2 examples

Answer 9

Regression tasks predict a numerical target variable from a set of features
* stock price forecasting
* weather temperature forecast

Question 10

Predictive clustering

What is predictive clustering

give 1 example

Answer 10

Predictive clustering predicts with the intention to assign class labels (predicting a target)
* fraud detection

Question 11

Descriptive Clustering

What is descriptive clustering

give 2 examples

Answer 11

The clusters are representing different groups formed in data without the intention of predicting a target.
* grouping plant data
* pattern mining

Question 12

Association rule mining

What is association rule mining

Give 2 examples

Answer 12

A rule-based task for discovering interesting relations between variables
* market basket anaylsis
* online shopping

Question 13

Sub-group discovery

What is subgroup discovery

give 2 examples

Answer 13

Technique that discovers interesting associations among different variables, with respect to a property of interest
* detection of risk groups with disease
* finding patterns in traffic accidents

Question 14

Supervised vs Unsupervised

What is supervised learning

Answer 14

In supervised learning tasks, we provide a traning set of examples: instances, labelled with the true target value.

Question 15

Supervised vs Unsupervised

What is Unsupervised learning

Answer 15

In unsupervised the data is unlabelled

Question 16

Name two Supervised and Predictive models

Answer 16

1. Classification
2. Regression

Question 17

Name one Supervised learning descriptive model

Answer 17

1. Subgroup discovery

Question 18

Name one unsupervised learning predictive model

Answer 18

1. Predictive clustering

Question 19

Name two unsupervised learning descriptive models

Answer 19

1. Descriptive clustering
2. association rule discovery

Question 20

What are Models?

Answer 20

Models are what is being learned from the data, in order to solve a given task

Question 21

How does a model of regression look like?

Equasion with Yi, Xi, ei

Answer 21

Yi = f(Xi+B)+ei
where Yi, Xi, ei are the target, features, and noise of specific instance i, and B and f are model paramters and model function

Question 22

What are the two ways machine learning models can be distinguished

Answer 22

1. Main intuition
2. Modus operandi (mode of operations)

Question 23

Main intuition

What are geometric models

Answer 23

using geometrical concepts. shit like linear tranformations, distance metrics, seperating hyperplanes

Question 24

Main intuition

what are Probabilistic models

Answer 24

aim for reducing uncertanty using probability distributions

Question 25

# Main intuition what are Logical Models

Answer 25

defined in terms of easily interpretable logical expressions

Question 26

# Second Categorization (modus operandi) What are gouping models

Answer 26

dividing the instance space into segments in each segment a very simple model is learnt

Question 27

# Modus operandi What are Grading models

Answer 27

learning a single, global model over the instance space

Question 28

# Geometric models Instance Space?

Answer 28

the set of all possible instances, whether they are present in the data set or not

Question 29

# Geometric models Distances?

Answer 29

distance between two points * Euclidean distance for two points. (could work in multiple dimentions)

Question 30

# Geometric models What are Hyper-Planes

Answer 30

a decision boundary that divides the input space into two or more regions, each corresponding to a different class or output label

Question 31

# Probabilitic models What is the bayes rule

Answer 31

P(Y|X)=P(X|Y)P(Y) P(X)

Question 32

# probabilitic models What is *posterior*

Answer 32

posterior = (likelihood x prior)/evidence

Question 33

# Proabilitic models P(Y|X) P(X|Y) P(Y) P(X)

Answer 33

P(Y|X) = posterior dist. P(X|Y) = likelihood prob. P(Y) = prior P(X) = evidence

Question 34

# Logical models what is meant by **Declerative**?

Answer 34

Declerative: models of this type can be earlisy translated into ruls that are understandable by humans * Rules can be organised into a feature tree

Question 35

# Grading vs Grouping Difference between grouping models and grading models. | Give 3 points for each

Answer 35

Grouping models * break up the instance space into **groups of segments** * grouping models have a **fixed and finite resolution** * cannot distingiush between indivual instances beyond this resolution Grading models * Do not work based on the notion of segments * they form one **global model **over the instance space * **infinite resolution**

Question 36

Give 1 example of a grouping model

Answer 36

1. Decision trees

Question 37

Give two examples of a grading model

Answer 37

1. Linear regression 2. Linear classifiers

Question 38

# Training vs inference What is the Training phase

Answer 38

Training is the process of creating a machine learning model that has learned ot perform a task using a training set of numerous data points.

Question 39

What is inference phase

Answer 39

Inference is the process of using a machine learning model to perform the task on a new data point.

Question 40

what are Features

Answer 40

Kind of measurement taht can be easily performed on any instance

Question 41

what if the data is not in the form you want?

Answer 41

1. Feature construction 2. Discretisation (numerical into categorical) 3. Feature transformation (project data into a new space) 4. Feature selection (removing redudntant features)