SUPERVISED MACHINE
LEARNING TECHNIQUES
TARA RAM
MCA 6th SEM
1.
Introduction
Terminology, type of supervised learning, Common Algorithms
& it’s applications
2
Instructions
3
▸ Supervised machine learning is subfields
machine learning.
▸ Supervised learning uses patterns to predict
label values on additional unlabeled data.
▸ Supervised learning is to use historical data
to predict statistically likely future events.
Cont..
4
▸ Supervised learning, in the context of
artificial intelligence (AI) and machine
learning, is a type of system in which both
input and desired output data are provided.
▸ In supervised learning, the system tries to
learn from the previous examples and
labeled data that are given.
Learning Process
5
Training Data
Machine learning
Algorithm
ClassifierNew Data Prediction
Supervised Vs Unsupervised Machine Learning Technique
Based On SML UML
Input Data
Trained using
labelled data
Trained using
unlabelled data
Accuracy of the
Result
More accurate and
reliable
Less acurate and
reliable
Number of classes Known Unknown
Real Time Learning
Learning tasks
place off-line
Learning takes
place in real time.
6
Terminology
7
▸ Data
▸ Problem solving tools
▸ Combinations of computer science and
engineering and statistics
▸ Optimize performance criteria using past
experience
Types of Supervised Machine
8
Supervised Machine Learning
Algorithms
Classification Supervised Machine
Learning Algorithms
Regression Supervised Mahine
Learning Algorithms
Classification Supervised Machine Learning Algorithms
▸ Classification algorithms
are used to classify a
records.
▸ A classification problem
is when the output
variable is a category or a
group.
9
Regression Supervised Machine Learning Technique
▸ Regression Algorithms are used to calculate
numeric values.
▸ A regression problem is when the output
variable is a real value, such as “Rupees” or
“height.”
10
Comman Algorithms
11
▸ Linear Regression
▸ Decision Trees
▸ Logistic Regression
▸ Naïve Bayes classifier
Linear Regression
12
▸ It is used to estimate real values (cost of houses,
number of calls, total sales etc.) based on
continuous variable(s).
▸ Establish relationship between independent
and dependent variables by fitting a best
line.
▸ This best fit line is known as regression
line.
Cont..
13
▸ A Standard and simple mathematical
techniques for predicting numeric
outcomes.
▸ Oldest and most widely predictive model
Decision Trees
▸ It is mostly used for classification problems.
▸ Decision trees classify instances or examples
by starting at the root of the tree and moving
through it until a leaf node
▸ Decision tree is a classifier in the form of a
tree structure.
14
Decision Tree
15
Age
Young Middle Old
Has_Job Own_House Credit_Rating
True false True False Fair Good
Yes No Yes No No Yes
Excellence
Yes
Logistic Regression
▸ It is used to estimate discrete values (Binary values
like 0/1, yes/no, true/false) based on given set of
independent variable(s).
▸ It predicts the probability of occurrence of an event
by fitting data to a logit function.
▸ Its output values lies between 0 and 1 (as
expected).
16
Naïve Bayes classifier
▸ Naïve Bayes Classifier technique based
on Bayes theorem.
▸ Naïve Bayesian model is easy to build.
▸ Naïve Bayes Classifier is used in large
data set.
17
Applications
▸ Speech Recognition
▸ Effective Web Search
▸ Machine Translation(Languages
translation)
▸ Email Spam Filtering
▸ Fraud Detection
▸ Medical Diagnosis
18
Cont..
▸ Stock Market Analysis
▸ Structural Health Monitoring
▸ Image Search (Similarity)
▸ Recommendation System
19
20
THANKS!
Tara Ram
MCA 6th SEM

Supervised Machine Learning Techniques

  • 1.
  • 2.
    1. Introduction Terminology, type ofsupervised learning, Common Algorithms & it’s applications 2
  • 3.
    Instructions 3 ▸ Supervised machinelearning is subfields machine learning. ▸ Supervised learning uses patterns to predict label values on additional unlabeled data. ▸ Supervised learning is to use historical data to predict statistically likely future events.
  • 4.
    Cont.. 4 ▸ Supervised learning,in the context of artificial intelligence (AI) and machine learning, is a type of system in which both input and desired output data are provided. ▸ In supervised learning, the system tries to learn from the previous examples and labeled data that are given.
  • 5.
    Learning Process 5 Training Data Machinelearning Algorithm ClassifierNew Data Prediction
  • 6.
    Supervised Vs UnsupervisedMachine Learning Technique Based On SML UML Input Data Trained using labelled data Trained using unlabelled data Accuracy of the Result More accurate and reliable Less acurate and reliable Number of classes Known Unknown Real Time Learning Learning tasks place off-line Learning takes place in real time. 6
  • 7.
    Terminology 7 ▸ Data ▸ Problemsolving tools ▸ Combinations of computer science and engineering and statistics ▸ Optimize performance criteria using past experience
  • 8.
    Types of SupervisedMachine 8 Supervised Machine Learning Algorithms Classification Supervised Machine Learning Algorithms Regression Supervised Mahine Learning Algorithms
  • 9.
    Classification Supervised MachineLearning Algorithms ▸ Classification algorithms are used to classify a records. ▸ A classification problem is when the output variable is a category or a group. 9
  • 10.
    Regression Supervised MachineLearning Technique ▸ Regression Algorithms are used to calculate numeric values. ▸ A regression problem is when the output variable is a real value, such as “Rupees” or “height.” 10
  • 11.
    Comman Algorithms 11 ▸ LinearRegression ▸ Decision Trees ▸ Logistic Regression ▸ Naïve Bayes classifier
  • 12.
    Linear Regression 12 ▸ Itis used to estimate real values (cost of houses, number of calls, total sales etc.) based on continuous variable(s). ▸ Establish relationship between independent and dependent variables by fitting a best line. ▸ This best fit line is known as regression line.
  • 13.
    Cont.. 13 ▸ A Standardand simple mathematical techniques for predicting numeric outcomes. ▸ Oldest and most widely predictive model
  • 14.
    Decision Trees ▸ Itis mostly used for classification problems. ▸ Decision trees classify instances or examples by starting at the root of the tree and moving through it until a leaf node ▸ Decision tree is a classifier in the form of a tree structure. 14
  • 15.
    Decision Tree 15 Age Young MiddleOld Has_Job Own_House Credit_Rating True false True False Fair Good Yes No Yes No No Yes Excellence Yes
  • 16.
    Logistic Regression ▸ Itis used to estimate discrete values (Binary values like 0/1, yes/no, true/false) based on given set of independent variable(s). ▸ It predicts the probability of occurrence of an event by fitting data to a logit function. ▸ Its output values lies between 0 and 1 (as expected). 16
  • 17.
    Naïve Bayes classifier ▸Naïve Bayes Classifier technique based on Bayes theorem. ▸ Naïve Bayesian model is easy to build. ▸ Naïve Bayes Classifier is used in large data set. 17
  • 18.
    Applications ▸ Speech Recognition ▸Effective Web Search ▸ Machine Translation(Languages translation) ▸ Email Spam Filtering ▸ Fraud Detection ▸ Medical Diagnosis 18
  • 19.
    Cont.. ▸ Stock MarketAnalysis ▸ Structural Health Monitoring ▸ Image Search (Similarity) ▸ Recommendation System 19
  • 20.