Week 1

Question 1

What is ‘Machine Learning’

Answer 1

a subset of artificial intelligence that enables systems to learn from data, identify patterns, and make decisions with minimal human intervention. This process is similar to how humans learn from their experiences.

Question 2

What are the 3 foundations of Machine Learning?

Answer 2

data, algorithms, and model

Question 3

What is Data?

Answer 3

data fuels machine learning. It encompasses different forms like numbers, text, images, or audio.

Question 4

What is Algorithms?

Answer 4

an algorithm is the logical procedure used to process data and learn patterns to make predictions or decisions.

Question 5

What is Models?

Answer 5

models are the end products of the machine learning process. It represents what the algorithm has learned from the data.

Question 6

What are the 3 types of Machine Learning?

Answer 6

Supervised Learning, Unsupervised Learning, and Reinforcement Learning.

Question 7

What is ‘Supervised Learning’?

Answer 7

in supervised learning, the algorithm is trained using a dataset containing input features and their corresponding output labels. This is similar to a teacher providing examples with answers to a student, who learns to predict answers for new examples.

Question 8

What are the 2 types of ‘Supervised Learning’

Answer 8

Regression and Classification

Question 9

What is Regression?

Answer 9

predicts a continuous numerical value.

Question 10

What is Classification?

Answer 10

predicts a categorical value.

Question 11

What is ‘Unsupervised Learning’

Answer 11

are trained on datasets with only input features provided. The algorithm must discover patterns or structures within the data. It resembles a learner organizing classmates without instructions. The learner observes similarities, such as clothing or backpacks, and forms groups based on patterns they discover, just as an algorithm finds structure in unlabeled data.

Question 12

What are the 2 types of ‘Unsupervised Learning’

Answer 12

Clustering and Dimensionality Reduction.

Question 13

What is Clustering?

Answer 13

groups similar data points together.

Question 14

What is ‘Dimensionality Reduction’

Answer 14

is the process of simplifying a dataset by reducing the number of features (or dimensions) while still keeping the most important information.

Question 15

What is ‘Reinforcement Learning’

Answer 15

reinforcement learning algorithms learn through interaction with an environment, receiving rewards or penalties based on their actions.
The algorithm learns to improve decisions based on feedback.

Question 16

What are the 5 steps in making ‘Machine Learning Models’

Answer 16

1. Data Collection and Preparation
2. Algorithm Selection
3. Model Training
4. Model Evaluation
5. Model Deployment

Question 17

Give examples of ‘Sypervised Learning Models’

Answer 17

Linear Regression, Logistic Regression, Decision Trees, and Random Forest

Question 18

What is ‘Linear Regression’

Answer 18

predicts a continuous output based on input features.

ex. predicting the cost of a new house based on house’s size.

Question 19

What is ‘Logistic Regression’

Answer 19

classifies data into discrete categories based on input features.

ex. classifying if an email is span or not spam, based on the presence of specific keywords.

Question 20

What is ‘Decision Trees’

Answer 20

makes decisions by splitting the data based on feature values.

ex. predicting if a new customer will purchase a product based on his age and income.

Question 21

What is ‘Random Forest’

Answer 21

a group of decision trees that improves accuracy and reduces overfitting.

Question 22

Give examples of ‘Unsupervised Learning Model’

Answer 22

K-Means Clustering, Hierarchial Clustering, and Principal Component Analysis(PCA)

Question 23

What is ‘K-Means Clustering’

Answer 23

divides data into non-overlapping clusters based on similarity.

ex. groups customer into distinct clusters based on their spending patterns.

Question 24

What is ‘Hierarcial Clustering’

Answer 24

builds a tree-like hierarchy of clusters.

Question 25

What is the difference between ‘K-Means Clustering’ and ‘Hierarchy Clustering’

Answer 25

K-Means Clustering: Groups data into a fixed number of clusters (
𝐾
K) by assigning points to the nearest cluster center. It’s fast and works well for large datasets but requires you to choose the number of clusters in advance.

Hierarchical Clustering: Creates a tree-like structure of clusters by merging or splitting data points step-by-step. It’s better for smaller datasets and doesn’t require choosing the number of clusters upfront, but it’s slower and harder to use for large data.

Question 26

What is ‘Principal Component Analysis’

Answer 26

reduces the dimensionality of data while retaining as much variance as possible.

Question 27

Give examples of ‘Reinforcement Learning Models’

Answer 27

Q-Learning and Deep Q-Networks (DQN).

Question 27

What is ‘Q-Learning’

Answer 27

bases its actions in a specific situations to maximize its rewards over time.

Question 28

What is ‘Deep Q-Networks’

Answer 28

an advanced version of Q-Learning that can handle more complex and large-scale environments, such as video games.

Question 29

What is ‘Data Analytics’

Answer 29

the process of examining data to extract insights and make informed decisions.

Question 29

What are the 4 main types of Data Analytics

Answer 29

Descriptive Analytics. Diagnostic Analysis, Predictive Analytics, and Prescriptive Analytics

Question 30

What is ‘Descriptive Analytics’

Answer 30

focuses on summarizing and describing past data. It helps to understand what has happened.

Question 31

What is ‘Diagnostic Analytics’

Answer 31

explores the underlying causes of events or trends. It helps to understand why something happened.

Question 32

What is ‘Predictive Analytics’

Answer 32

uses historical data to predict future outcomes. It helps to anticipate what will happen.

Question 33

What is ‘Prescriptive Analytics’

Answer 33

suggests optimal actions based on data analysis and predictions. It helps determine what should be done.

Question 34

What is ‘Exploratory Data Analysis’?

Answer 34

Exploratory Data Analysis (EDA) is the process of analyzing and visualizing data to understand its patterns, trends, and relationships before applying formal modeling techniques. It’s like taking a first look at your data to see what’s interesting or unusual.

It involves ‘Summarizing Data’, ‘Visualizing Data’, ‘Checking Relationsips’ and ‘Spotting Issues’

EDA is like getting to know your data before diving into complex analysis. It’s the “exploration” phase where you clean up, visualize, and make sense of the data to guide your next steps.

Question 34

What are the 6 steps in Data Analytics?

Answer 34

1. Define the Problem Statement.
2. Collect the data.
3. Clean the Collected Data.
4. Analyze and Interpret the Cleaned Data.
5. Visualize the Interpreted Data.
6. Present the Analysis Results.

Question 35

What are the key characteristics of ‘Descriptive Analytics’

Answer 35

Data Summarization, Trend Identification, and Performance Metrics.

Question 36

What are the key characteristics of ‘Diagnostic Analytics’

Answer 36

Root Cause Analysis, Data Drill-Down, Correlation and Causality, and Hypothesis Testing.

Question 37

What are the key features of ‘Prescriptive Analytics’

Answer 37

Actionable Recommendations, Optimization, Predictive Models, Simulation and Scenario Analysis, and Complex Algorithms and Machine Learning.

Question 38

What are the key features of ‘Predictive Analytics’

Answer 38

Forecasting, Pattern Recognition, Probability Scores, Use of Advanced Algorithms, Scenario Testing.

Question 39

What are the techniques used in ‘Predictive Analytics’

Answer 39

Regression Analysis, Time Series Analysis, Machine Learning Models, Classification Models, and Clustering.

Question 40

What is ‘Regression Analysis’

Answer 40

a statistical method usd to determine the relationship between a dependent variable and one or more independent variable.