Notesale: Turn your study into money

Document Preview

Extracts from the notes are below, to see the PDF you'll receive please use the links above

---

Artificial Intelligence Full Course ,Artificial
Intelligence Tutorial for Beginners
This course provides an introduction to artificial intelligence, discussing its history, different
domains and concepts, and how deep learning is used to solve real-world problems
...

This covers the history of artificial intelligence, the different domains and concepts
involved, how artificial intelligence came into existence, the limitations of machine
learning, and how deep learning is needed
...
Finally, covers
the next module, natural language processing
...
Recent advances in computing power and algorithms have made it
possible for AI to be implemented in our daily lives more efficiently
...
AI is rapidly growing both as a field of study and also as an economy
...
AI has become so important that it is used by even the smallest of
businesses, like Netflix
...
Artificial general intelligence, or strong AI, is still far away from being
achieved, but if it were to happen, it would be a major milestone in human history
...
Python is considered to be the best
language for AI development, and R is also a popular choice
...

Python is a versatile, easy-to-use programming language that is gaining popularity in
the field of artificial intelligence
...

Machine learning is a subset of artificial intelligence that uses algorithms to learn
automatically and improve with experience
...

The difference between an algorithm and a model is that an algorithm maps all the
decisions that a model is supposed to take based on the given input, while a model
will use the machine learning algorithm in order to draw useful insights from the
input and give you an outcome that is very precise
...
So, in the
same example, height would be the response variable
...
This is the variable that you're
trying to predict by using the predictor variables
...
Next, we have something known as the training data
...
So, training data is basically the data that I used to create the machine
learning model
...
So splitting the data into two parts is also
known as data splicing
...
One you'll
Data gathering is one of the most time-consuming steps in machine learning, and if
you have to go manually and collect the data, it will take a lot of time
...
All you need to do
is web scraping where you just have to go ahead and download data
...
So if you're a beginner in machine learning,
don't worry about data gathering and all of that
...

Supervised learning is a technique in which a machine is taught by using data that is
well-labeled
...

In supervised learning, a training data set is labeled with information about how
things look like, such as images of Tom and Jerry
...
In
unsupervised learning, the machine learning algorithm is not fed with any labeled
data, but is instead allowed to learn by using unlabeled data
...

Machine learning consists of three main types of learning: supervised learning,
unsupervised learning, and reinforcement learning
...
There are
three types of problems that can be solved using machine learning: regression,
classification, and clustering
...


This discusses various types of artificial intelligence and how they can be used to solve
various types of problems
...
It also discusses the
decision tree algorithm and how it can be used to build a decision tree
...


•

Artificial intelligence can be used to solve classification problems, regression
problems, and clustering problems
...

Linear regression is a supervised learning algorithm used to predict a continuous
dependent variable, y, based on the values of independent variable, x
...

In this video, Edureka instructor Michael Kennedy demonstrates linear regression
with a data set of weather conditions recorded on different days around the world
...
He also discusses the
warning message and explains that the main aim of this demo is weather forecast
...
By splitting the
data set into training and testing sets, and importing the appropriate linear
regression class, the model is trained
...

This video explains how to use a regression algorithm to predict the percentage
score of a test data set
...

Logistic regression is a method used to predict a dependent variable, y, given an
independent variable, x, such that the dependent variable is a categorical variable,
meaning that the output is a categorical variable
...

Logistic regression is used for predicting the probability of an outcome being one or
zero, using the equation Pr(X = 1) = beta naught + beta one into X
...

The decision tree algorithm is a supervised machine learning algorithm that is easy
to understand
...
Branches between nodes are represented by arrows, and the algorithm is
run by passing the data through the tree until the terminal node is reached
...
The steps involved
in using this algorithm are as follows: (1) select the best attribute, (2) assign this
attribute as a decision variable for the root node, (3) for each value of the decision
variable, build a descendant node, and (4) assign classification labels to the leaf
nodes
...
The best attribute is the one that separates the data into
different classes most effectively
...
The highest information gain
variable will be used to split the data at the root node
...

Entropy is calculated for the right-hand side and is found to be zero, indicating no
uncertainty
...
When the road type is steep, the speed of
the car can be slow or fast, indicating that the information is not specific to any
particular road type
...
entropy is calculated for the parent node, weighted average is calculated for
the children, and information gain is calculated for each predictor variable
...
The information gain for the road type variable is 0
...
The
information gain for obstruction is zero, meaning that the obstruction variable has
no impact on the decision tree
...

In a random forest, multiple decision trees are built and glued together to create a
more accurate and stable prediction
...
Random forest is more accurate than
decision trees when it comes to predicting new data, because overfitting (
memorizing the training data) is reduced
...
He also discusses
how reinforcement learning is a different kind of machine learning that helps agents learn
how to achieve their goals in an unknown environment
...
First, they select two or three variables at random to be used in each node of
the decision tree, and then calculate information gain and entropy for each
...

Finally, they go back to step one and create a new decision tree based on a subset of
the original variables
...
Finally, they evaluate the model's accuracy by using the out-of-bag data set
...
First, a bootstrap data set is created to allow for accurate predictions
...
This process is
repeated hundreds of times until a model is created
...

The K nearest neighbor algorithm is a supervised classification algorithm that
classifies a new data point into the target class or the output class, depending on the
features of its neighboring data points
...
It is based on feature similarity with its neighboring
data points and is non-parametric
...


•

•

•

•

•

•

•

•

In this video, Edureka instructor Alan See explains the basics of three classification
algorithms: KNN, SVM, and support vector machines
...

This video discusses how to use various classification algorithms with Python
...
Next,
various algorithms are implemented and tested on the data
...

The video discusses the importance of visualization in machine learning, and goes
on to explain the use of box plots, histograms, and scalers
...

This video covers the use of logistic regression, decision trees, and support vector
machines in classification
...
The decision tree
classifier was more accurate on the training data set but performed less well on the
testing data set
...

The K-means clustering algorithm is a unsupervised learning algorithm used for
grouping similar elements or data points into clusters
...

The elbow method is a simple method used to find the optimum k value for a
particular problem
...
As K gets larger,
the error decreases, indicating that more clusters result in smaller distortion
...
This method uses the squared sum of errors to determine the distortion, and the
most optimal K-means K value is found by finding the point at which the distortion decreases
abruptly
...
In this
video, we use a sample image from the scikit-learn data set to demonstrate the elbow
method
...
It also discusses how
reinforcement learning is a different kind of machine learning that helps agents learn how to
achieve their goals in an unknown environment
...
The video explains how a reward matrix and equitable Q
matrix are used to determine the agent's current state and future rewards
...


•

The reinforcement learning agent in a video game, such as in Counter Strike, tries to
maximize its reward by taking the best action given its current state and environment
...
These concepts, such as action, state, reward, and gamma, will
be covered in more detail in later slides
...
He then demonstrates an example of how to choose a policy for a
problem using the greedy strategy and an example of how to choose a policy using the
exploitation strategy
...
He then
demonstrates the algorithm Q-learning, which is an important reinforcement learning
algorithm
...

This video explains the basics of artificial intelligence, including how it works and how to
create an agent that can learn from experience
...

Gamma is used to control the agent's exploration and exploitation
...
It then goes on to show how to do the same thing in
Python using the NumPy and R libraries
...
The video then goes on to show how to train the AI system by running it through
10,000 iterations, and how to test the system by randomly selecting a state and trying to
reach the goal state, which is room number five
...
However, it is not
capable of handling highdimensional data
...

Artificial intelligence is restricted in its ability to be used for image recognition because
images have a lot of pixels and they have a lot of high dimensional data
...
Deep learning mimics the way our brain
functions and can learn to focus on the right features by itself, which requires very little
guidance from the programmer
...
Deep learning consists of a network of artificial neurons, known as
artificial neural networks, which work exactly like how our brain operates
...

In this video, the instructor explains how weightage is used in the calculation of an
activation function
...

A multilayer perceptron has the same structure of a single layer perceptron, but with one or
more hidden layer
...
Back propagation is a way to
update the weights in order to reduce the error
...
First, they calculate the error, which reveals where the model
is inaccurate
...
If the error remains high, they stop updating the
weights and find the global loss minimum, at which point they stop
...
He
explains how neural networks work and how they can be used to predict stock prices
...
Finally, he discusses the model architecture parameters for an artificial
intelligence system, including the number of neurons in each hidden layer, the bias
dimension, and the cost function
...
Gradient descent is used to optimize
the work of a feed forward network
...


This video explains how deep neural networks work and how they can be used to predict
stock prices
...

•

•

•

•

•

•

The video covers the steps necessary to train a neural network, including data
preparation, data splicing, and scaling
...

The video discusses the model architecture parameters for an artificial intelligence
system, including the number of neurons in each hidden layer, the bias dimension,
and the cost function
...

In this video, Edureka instructor Kirill Eremenko explains the basics of deep learning,
including the role of neural networks, optimizers, and initializers
...

In this video, deep learning is demonstrated by comparing the predicted values of a
model to the actual observed targets, which is stored in y
...
The model is then trained on the test
data, and its prediction is compared to the actual values
...

Text mining or text analytics is the process of deriving meaningful information from
natural language text
...
NLP is a part of text mining which helps machines
understand the data in the form of zeroes and ones
...
Applications of text mining and natural language processing
include spam detection, predictive typing, and sentimental analysis
...
It explains that tokenization breaks a sentence into
words, stemming reduces words to their base form, and lemmatization links words
back to their lemma
...


•

•

•

In this tutorial, Edureka teaches how to perform natural language processing by
using the NaiveBayesClassifier, which is a library that contains all the necessary
functions to perform the task
...
The classifier was able to accurately
identify which reviews were positive or negative
...
The curriculum includes supervised
and unsupervised algorithms, statistics and time series, deep learning, and Spark
...

This video provides a comprehensive overview of artificial intelligence, covering the
basics of programming, data, and machine learning

---