Scribd
Upload
12 views

Classification-Basic Concepts

The document outlines the basic concepts of classification in data analysis, emphasizing its role in creating classifiers that categorize data into target classes. It describes the two-step process of building a classification model and using it to classify new data, highlighting the distinction between supervised and unsupervised learning. Additionally, it discusses the importance of measuring learning accuracy through test sets to ensure the model's effectiveness in predicting unknown class labels.

Uploaded by

umakanthn
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
12 views

Classification-Basic Concepts

The document outlines the basic concepts of classification in data analysis, emphasizing its role in creating classifiers that categorize data into target classes. It describes the two-step process of building a classification model and using it to classify new data, highlighting the distinction between supervised and unsupervised learning. Additionally, it discusses the importance of measuring learning accuracy through test sets to ensure the model's effectiveness in predicting unknown class labels.

Uploaded by

umakanthn
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 13

Classification-Basic

Concepts

By,
Umakanth N
Session Outcomes

 Identify situations which demand use of


classification
 Decide to use clustering or classification according
to the given problem

February 10, 2025 Classification-Basic Concepts 2


Classification-Basic
Concepts
 It is a form of data analysis that extracts models
describing important data classes. Such models
are called Classifiers
 It assigns the items in a collection to the target
categories or classes
 Eg., Bank-Loan-Classifier should be able to
identify the credit-risk of applicant as Low,
Medium or High
 Classification as a Two-Step Process:
 Build classification model based on previous data
 If models’ accuracy is acceptable, use the model to
classify new data
February 10, 2025 Classification-Basic Concepts 3
Classification - Example

February 10, 2025 Classification-Basic Concepts 4


Data Classification
 Predicting the categorical class labels. Eg.,
Safe / Risky for loan application data; Treatment
A / B / C for medical data.
 Can be represented by discrete values; but ordering
the values have no meaning
 Numeric prediction for the continuous valued
attributes
 Regression analysis is the statistical methodology,
which was used for numeric prediction

February 10, 2025 Classification-Basic Concepts 5


Data Classification
(Contd.,)
 Two Step Process:
 Learning Step (Training Phase)
 Classification model is constructed
 Classification step
 Model is used to predict the label for the given data

February 10, 2025 Classification-Basic Concepts 6


Data Classification: Learning
step
 Classification algorithm builds the classifier by
analyzing the training set made up of database
tuples and their associated class labels
 A tuple X is represented by n-dimensional
attribute vector, X = {x1, x2, ….. xn}; with
another class labeled attribute.
 Class labeled attributes are discrete and
unordered
 Data tuples can be referred as samples,
examples, instances, data points or objects (in
data classification terms)

February 10, 2025 Classification-Basic Concepts 7


Data Classification: Learning
step

February 10, 2025 Classification-Basic Concepts 8


Data Classification: Learning
step
 If the class label of each training tuple is provided
during the Learning of the classifier, is called
Supervised Learning.
 If class label is not known, then learning process
is called Unsupervised Learning (or clustering)
 Learning step can be viewed as the learning of
mapping or function, y = f (x) ie., predicts
associated class label y for given tuple x.
 These mappings can be represented in the form of,
 Classification rules,
 Decision trees,
 Mathematical formulae

February 10, 2025 Classification-Basic Concepts 9


Data Classification: Learning
step
 How to measure the Learning Accuracy?
 By using Test set; which has Test tuple and their
associated class label
 Accuracy of the classifier is the percentage of test
set tuples that are correctly classified by it.
 If accuracy is acceptable, then the model can be used
to classify the future data tuples for the unknown class
labels.

February 10, 2025 Classification-Basic Concepts 10


Data Classification:
Classification step

February 10, 2025 Classification-Basic Concepts 11


Test Yourself..!!!
 A college admission has data set for its 10 batch
of students. It wants to predict the placement of
students during the admission. What would be
the class labels for this?
 If a person age is less than 30, and he is working
the possibilities of purchasing a smart phone is
high else it is less. Is it a classifier problem? If so,
what are the attributes in this statement?
 Write the decision rule for this=>

February 10, 2025 Classification-Basic Concepts 12


? ?
. . ?
ie s
e r
Q u

You might also like