Scribd
Upload
4 views

Data Mining

Uploaded by

James Bacolod
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
4 views

Data Mining

Uploaded by

James Bacolod
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 24

Data

Mining
TECHNIQUES
Reported by: James Kyle Bacolod
Data Data mining techniques are
methods and algorithms used

Mining to analyze and extract


meaningful patterns or
TECHNIQUES insights from large datasets.
ASSOCIATION
CLASSIFICATION CLUSTERING REGRESSION
RULE MINING

ANOMALY
Data Mining DIMENSIONALITY
DETECTION REDUCTION
Techniques

NEURAL
TIME SERIES SEQUENTIAL
NETWORKS (DEEP TEXT MINING
ANALYSIS PATTERN MINING
LEARNING)
Classificatio
Classification is a supervised learning
n
technique where the goal is to predict
the categorical label (class) of new
data points based on a labeled
dataset.
Example: Medical Diagnosis

Problem: Classifying Patient Health


Conditions

• Scenario: A healthcare provider wants to


classify patients as having a specific
disease (e.g., diabetes) or not based on
their medical data.
Classificatio
n
Clustering
Clustering is an unsupervised learning
technique where the data is grouped
into clusters based on similarity without
predefined labels.

Example: Customer Segmentation for Marketing

• Problem: Grouping customers based on


purchasing behavior to create targeted
marketing campaigns.
Clustering(K Means
clustering)High
Spenders

Occasional
Buyers

Frequent
Buyers

befor after
e
Regression
Regression is used to predict a continuous
value based on input variables. It’s a
supervised learning technique used for
modeling relationships between dependent
and independent variables.

Example: Predicting Student Academic


Performance
• Problem: Estimating the final exam
scores of students based on various
input features such as attendance, study
hours, and previous test scores.
Regression
( Multi Linear
Regression)
Association rule
Association rule mining finds hidden

mining
relationships between variables in large
datasets, most commonly used for
market basket analysis.

Example: Market Basket Analysis

• Problem: Identifying items that are frequently


purchased together in a grocery store.
Association rule
(Apriori algorithm)
mining
The Apriori algorithm is used to find
frequent itemsets (e.g., combinations
of items that are frequently bought
together) and generate association
rules. For example, a rule might be "If
a customer buys bread, they are 70%
likely to buy butter."
Anomaly
Anomaly detection identifies unusual
detection
patterns or data points that don’t conform
to expected behavior. It is used to detect
rare events or deviations in data.

Example: Fraud Detection in Credit Card


Transactions

• Problem: Detecting fraudulent credit card


transactions that deviate from a customer’s
normal purchasing behavior.
Anomaly
(Isolation Tree)
detection
Dimensionality
Dimensionality reduction reduces the

reduction
number of input variables (features)
while retaining as much information as
possible, simplifying the model and
speeding up computation.

Example: Image Compression

• Problem: Reducing the size of an image while


preserving essential features for storage and
analysis.
Dimensionality
reduction
Neural networks
Neural networks, inspired by the structure
of the brain, are powerful models that can
identify complex patterns in large datasets.
Deep learning involves neural networks with
many layers (deep architectures).

Example: Image Recognition

• Problem: Identifying objects in an image (e.g.,


recognizing cats and dogs).
Neural networks

(Convolutional Neural
Network)
Time series analysis
Time series analysis deals with data
points collected or recorded at regular
time intervals. The goal is to model and
predict future values based on past
observations.

Example: Stock Price Prediction

• Problem: Predicting the future price of a stock


based on its historical price data.
Time series analysis

ARIMA (AutoRegressive Integrated Moving


Average)
Text
Mining
Text mining extracts meaningful insights
and patterns from unstructured text data.
It’s commonly used in natural language
processing (NLP) tasks.

Example: Sentiment Analysis of Customer Reviews

• Problem: Determining whether customer


reviews for a product are positive, negative, or
neutral.
Text
Mining

( NPL - (TF-
IDF) )
Sequential pattern
Sequential pattern mining identifies
patterns in sequences of data, such as
event logs or transactional data that
follow a specific order.

Example: E-Commerce Purchase Patterns

• Problem: Identifying sequences of products that


customers frequently buy over time in an
online store.
Sequential
pattern

PrefixSpan
algorithm
Thank You!

You might also like