Machine learning and data mining
Introduction - correct answer-
What is data mining? - correct answer-Non-trivial extraction of implicit,
previously unknown and potentially useful information from data
Exploration & analysis, by automatic or semi-automatic means, of large
quantities of data in order to discover meaningful patterns
What kind of data to mine? - correct answer-• Relational Databases
• Data Warehouses
• Transactional Databases
• Advanced Database Systems
Challenges of Data Mining - correct answer-- Scalability
- Dimensionality
- Complex and Heterogeneous Data
- Data Quality
- Data Ownership and Distribution
- Privacy Preservation
- Streaming Data
What is machine learning? - correct answer-Field of study that gives
computers the ability to learn without being explicitly programmed.
challenges of machine learning - correct answer-bad data, noisy
missing values
irrelevant features
model selection
When to use machine learning - correct answer-Human expertise does not
exist
Humans can't explain their theory or expertise
Where models must be customised
where models are based on huge amount of data that
Goal of Machine Learning - correct answer-to build computational models with
high prediction and generalization capabilities
,steps to solve machine learning problem - correct answer-Data gathering
Data pre-processing
Feature engineering
Algorithm selection & training
Making predictions
data gathering - correct answer-process of obtaining existing, readily available
data
manual labelling of supervised learning
data preprocessing - correct answer-Data consolidation
Data cleaning
Data transformation
Data reduction
what is feature engineering - correct answer-converting raw data into a set of
features/quantitative metrics to put into a model.
# of calls, etc
insight from deterministic finate antonoma (recursively generate features
automatically).
What is a feature? - correct answer-A characteristic of the product
A feature is an individual measurable property of a phenomenon being
observed
algorithmic selection & training - correct answer-Selecting the right machine
learning model
making predictions - correct answer-Evaulate the model
Types of Machine Learning - correct answer-Supervised
Unsupervised
Reinforcement
What is supervised learning? - correct answer-Supervised learning is the
machine learning task of inferring a function from labeled training data. The
, training data consist of a set of training examples. In supervised learning,
each example is a pair consisting of an input object (typically a vector) and a
desired output value (also called the supervisory signal).
What is unsupervised learning? - correct answer-Unsupervised learning is the
machine learning task of inferring a function to describe hidden structure from
unlabeled data. Since the examples given to the learner are unlabeled, there
is no error or reward signal to evaluate a potential solution. This distinguishes
unsupervised learning from supervised learning and reinforcement learning.
What is reinforcement? - correct answer-Reinforcement learning is a machine
learning training method based on rewarding desired behaviours and/or
punishing undesired ones. In general, a reinforcement learning agent is able
to perceive and interpret its environment, take actions and learn through trial
and error.
What is classification in machine learning? - correct answer-A common job of
machine learning algorithms is to recognize objects and being able to
separate them into categories. This process is called classification, and it
helps us segregate vast quantities of data into discrete values, i.e. :distinct,
like 0/1, True/False, or a pre-defined output label class.
ETHICS OF MACHINE LEARNING - correct answer-
GDPR (General Data Protection Regulation) - correct answer-1. Big data
analytics must be fair.
2. Permission to process data.
3. Purpose limitation.
4. Holding on data.
5. Accuracy.
6. Individual rights and access to data.
7. Security measures and risk.
8. Accountability.
9. Controllers and processors.
Bias issues in DM/ML algorithms - correct answer-Algorithmic bias (feature or
model selection)
Data bias (biased or irrelevant data)
Interpretability/Transparency of DM/ML systems - (model bias)
Introduction - correct answer-
What is data mining? - correct answer-Non-trivial extraction of implicit,
previously unknown and potentially useful information from data
Exploration & analysis, by automatic or semi-automatic means, of large
quantities of data in order to discover meaningful patterns
What kind of data to mine? - correct answer-• Relational Databases
• Data Warehouses
• Transactional Databases
• Advanced Database Systems
Challenges of Data Mining - correct answer-- Scalability
- Dimensionality
- Complex and Heterogeneous Data
- Data Quality
- Data Ownership and Distribution
- Privacy Preservation
- Streaming Data
What is machine learning? - correct answer-Field of study that gives
computers the ability to learn without being explicitly programmed.
challenges of machine learning - correct answer-bad data, noisy
missing values
irrelevant features
model selection
When to use machine learning - correct answer-Human expertise does not
exist
Humans can't explain their theory or expertise
Where models must be customised
where models are based on huge amount of data that
Goal of Machine Learning - correct answer-to build computational models with
high prediction and generalization capabilities
,steps to solve machine learning problem - correct answer-Data gathering
Data pre-processing
Feature engineering
Algorithm selection & training
Making predictions
data gathering - correct answer-process of obtaining existing, readily available
data
manual labelling of supervised learning
data preprocessing - correct answer-Data consolidation
Data cleaning
Data transformation
Data reduction
what is feature engineering - correct answer-converting raw data into a set of
features/quantitative metrics to put into a model.
# of calls, etc
insight from deterministic finate antonoma (recursively generate features
automatically).
What is a feature? - correct answer-A characteristic of the product
A feature is an individual measurable property of a phenomenon being
observed
algorithmic selection & training - correct answer-Selecting the right machine
learning model
making predictions - correct answer-Evaulate the model
Types of Machine Learning - correct answer-Supervised
Unsupervised
Reinforcement
What is supervised learning? - correct answer-Supervised learning is the
machine learning task of inferring a function from labeled training data. The
, training data consist of a set of training examples. In supervised learning,
each example is a pair consisting of an input object (typically a vector) and a
desired output value (also called the supervisory signal).
What is unsupervised learning? - correct answer-Unsupervised learning is the
machine learning task of inferring a function to describe hidden structure from
unlabeled data. Since the examples given to the learner are unlabeled, there
is no error or reward signal to evaluate a potential solution. This distinguishes
unsupervised learning from supervised learning and reinforcement learning.
What is reinforcement? - correct answer-Reinforcement learning is a machine
learning training method based on rewarding desired behaviours and/or
punishing undesired ones. In general, a reinforcement learning agent is able
to perceive and interpret its environment, take actions and learn through trial
and error.
What is classification in machine learning? - correct answer-A common job of
machine learning algorithms is to recognize objects and being able to
separate them into categories. This process is called classification, and it
helps us segregate vast quantities of data into discrete values, i.e. :distinct,
like 0/1, True/False, or a pre-defined output label class.
ETHICS OF MACHINE LEARNING - correct answer-
GDPR (General Data Protection Regulation) - correct answer-1. Big data
analytics must be fair.
2. Permission to process data.
3. Purpose limitation.
4. Holding on data.
5. Accuracy.
6. Individual rights and access to data.
7. Security measures and risk.
8. Accountability.
9. Controllers and processors.
Bias issues in DM/ML algorithms - correct answer-Algorithmic bias (feature or
model selection)
Data bias (biased or irrelevant data)
Interpretability/Transparency of DM/ML systems - (model bias)
