Scribd
Upload
11 views

Introduction of Machine Learning

Uploaded by

AVINASH SAH
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
11 views

Introduction of Machine Learning

Uploaded by

AVINASH SAH
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 9

Introduction of machine learning

It is the field of studies that gives computers a capability to learn without


being explicitly program. Machine adopts to the user based on data.
Examples: Online shopping
Today, machine learning is one of the most common forms of artificial intelligence and often
powers many of the digital goods and services
Types of machine learning :
1. Supervised machine learning
2. Unsupervised machine learning
3.Semi-Supervised Machine Learning
3. Reinforcement learning
 Supervised machine learning:
 In supervised machine learning, algorithms are trained on labeled data sets that include tags describing
each piece of data.
 Supervised Learning algorithms takes place in two stages.
 first stage ,The teacher communicates the information to the students that the student is supposed to a
master. The students receives the information and understand it.during this stage the teacher has no
knowledge wheather the information is grasped by the students.
 Second stage, teacher asked the students a set of questions to find how much information has been
grasped by the students. based on these questions the student is tested,and the teacher informs the
students about his assessment.this type of leaening is called supervised learning.
Supervised machine learning is often used to create machine learning models used for prediction and
classification purposes.
Super vised learning has two methods:
Classification
The relationship between input and target variables is represented in the form of structure which is called a
classification.
Input attribute of classification algorithim are called independent variables.target attributes are called
dependant variables.
Learning algorithim takes the labelled dataset and start learning.after training set samples are processed and
model is generated.
Second stage,the constructed model is tested with test or unknown samples are assigned a lebel.
Some of the key algorithms of classification are:
1. Decision Tree
2. Random Forest
3. Support vector machine
4.Naïve Bayes
5. ANN & CNN
Regression:
It predict the continuous variables like price.in other words it is a number.
The regression model takes input x and generates a model in the form of a fitted line of the form y=f(x).
Unsupervised machine learning :
Unsupervised machine learning uses unlabeled data sets to train algorithms. In this process, the algorithm is
fed data that doesn't include tags, which requires it to uncover patterns on its own without any outside
guidance.
Unsupervised machine learning is often used by researchers and data scientists to identify patterns within
large, unlabeled data sets quickly and efficiently.

Cluster analysis and Dimensional reduction algorithm are examples of unsupervised algorithm.
Cluster objects based on its attributes.
Cluster Analysis:
Cluster analysis is the examples of unsupervised learning.it aims to group objects into disjoint clusters or
groups. cluster analysis cluster objects based on its attributes. All the data objects of the partitions are similar
in some aspect and vary from the data objects in the other partitions significantly.
Examples: detection of abnormal growth in medical image, determine clusters of signatures in a gene
database.in clustering algorithm takes a set of dogs and cats images and groups it as two cluster dog and cat.
Dimensionality Reduction:
It takes the higher dimension data as input and outputs the data in lower dimension by taking advantage of the
variance of the data.it is task of reducing the dataset with few features without losing the generality.
Semi-Supervised Machine Learning:
where the dataset has a huge collection of labelled data and
unlabelled data. Labelling is a costly process and difficult to perform
by the humans. Semi-supervised algorithms use unlabelled data by
assigning pseudo label.

Reinforcement learning:
Reinforcement learning uses trial and error to train algorithms and
create models. During the training process, algorithms operate in
specific environments and then are provided with feedback following
each outcome.
Reinforcement learning is often used to create algorithms that must
effectively make sequences of decisions or actions to achieve their
aims, such as playing a game or summarizing an entire text.
 Support Vector Machine

Support Vector Machine (SVM) is a supervised learning algorithm that takes a labelled
data as input and creates learning functions that can be used for classification of
unknown test data.

Though we say regression problems as well it’s best suited for classification. The main
objective of the SVM algorithm is to find the optimal hyperplane in an N-dimensional
space that can separate the data points in different classes in the feature space.
The hyperplane tries that the margin between the closest points of different classes
should be as maximum as possible. The dimension of the hyperplane depends upon the
number of features. If the number of input features is two, then the hyperplane is just a
line. If the number of input features is three, then the hyperplane becomes a 2-D plane.
It becomes difficult to imagine when the number of features exceeds three.
Let’s consider two independent variables x1, x2, and one dependent variable
which is either a blue circle or a red circle.
From the figure above it’s very clear that there are multiple lines (our
hyperplane here is a line because we are considering only two input features
x1, x2) that segregate our data points or do a classification between red and
blue circles.
Q. Suppose you are running a factory some sort of widget that requires steel
as a raw material. your cost are predominately human labor which is $20 per
hour for workers and the steel itself which runs $170 per ton. suppose your
revenue R is loosely modeled by the following equations
R(h,s) = 200. h2/3.s1/3 . If your budget is $20000 what is the maximum possible
revenue?
Advantages of SVM

 Svm can perform as linear and non-linear classifier.


 Svm can perform regression.
 The decision boundary can be constructed using a small set of
support vectors and can be constructed using less training samples.
but this factor is dependent on the nature of the given application.
 Svm works with higher dimensional data.
 Generally, svm classifiers are very robust and immune to the data
features or dimensions.
 Svm can be implemented in many applications such as object
recognition face recognition iris classification and pedestrian
recognition successfully.

You might also like