Introduction to-data-mining chapter 1

Introduction to Data Mining
Mahmoud Rafeek Alfarra
http://mfarra.cst.ps
University College of Science & Technology- Khan yonis
Development of computer systems
2016
Chapter 1

Outline
 Definition of Data Mining
 Data Mining as an Interdisciplinary field
 Process of Data Mining
 Data Mining Tasks
 Challenges of Data Mining
 Data mining application examples
 Introduction to RapidMiner

Definition of Data Mining
"Computers have promised us a source of wisdom but
delivered a flood of data."
"It has been estimated that the amount of information in
the world doubles every 20 months."
The Explosive Growth of Data: from terabytes to
petabytes
We are drowning in data, but starving for knowledge!

 Knowledge discovery in databases (data
mining) is
“The non-trivial process of identifying valid,
novel, potentially useful, and ultimately
understandable patterns in data”.

 Pattern is an arrangement of repeated parts.
 In a data table, a pattern is defined as a set of rows
that share the same values in two or more columns.
 Consider for example, the following table that
contains data about objects; shape, color, and weight.

WeightColorShapeRow #
100RedBox1->
200RedBox2->
300RedBox3->
400BlueBox4
400BlueCone5
In this table, we have 3 rows (row 1, 2 and 3) that share the same values
in two columns (Shape and Color). From this table, we can observe the following
patterns:
Most Boxes are Red.
We can represent Pattern as rule:
If Shape = Box then Color = Red.

 Valid: Discovered patterns should be true
on new data with some degree of certainty.
Generalize to the future (other data).
 Novel: Patterns must be novel (should not
be previously known).

 Actionable: patterns should potentially lead to
some useful actions.
 Understandable: Patterns must be made
understandable in order to facilitate a better
understanding of the underlying data.

Example: Credit Risk
A credit risk is the risk of default on a debt that may arise from a
borrower failing to make required payments.
In the first resort, the risk is that of the lender and includes lost principal
and interest, disruption to cash flows, and increased collection costs.

Is it valid?
The pattern has to be valid with respect to a certainty level (rule true for
the 86%)
Is it novel?
The value k should be previously unknown or obvious
Is it useful?
The pattern should provide information useful to the bank for assessing
credit risk
Is it understandable?

 Other definition of data mining:
“Is the process of extracting knowledge hidden from
large volumes of raw data. The knowledge must be
new, not obvious, and must be able to use it”.

Many people treat data mining as a synonym for
another popularly used term , knowledge Discovery
in Databases, or KDD. Alternatively, other view
data mining as simply an essential step in the
process of knowledge discovery in databases.

What is Data Mining?What is not Data Mining?
Certain names are more common in
certain US locations (O’Brien,
O’Rurke, O’Reilly … in Boston area)
Look up phone number in
phone
directory
Group together similar documents
returned by search engine according
to their context (e.g. Amazon
rainforest, Amazon.com,) information
about “Amazon”
Query a Web search engine

Data Mining and Business Intelligence
Increasing potential
to support
business decisions
End User
Business
Analyst
Data
Analyst
DBA
Decision
Making
Data Presentation
Visualization Techniques
Data Mining
Information Discovery
Data Exploration
Statistical Summary, Querying, and Reporting
Data Preprocessing/Integration, Data Warehouses
Data Sources
Paper, Files, Web documents, Scientific experiments, Database Systems

2016
Lecture 2

Outline
Definition of Data Mining
Data Mining as an Interdisciplinary field
Process of Data Mining
Data Mining Tasks
Challenges of Data Mining
Data mining application examples
Introduction to RapidMiner

Data Mining as an Interdisciplinary field
“Data mining is an interdisciplinary field bringing
together techniques from machine learning,
pattern recognition, statistics, databases, and
visualization to address the issue of information
extraction from large data bases”.

Data Mining
Database
Technology
Statistics
Other
Disciplines
Artificial
Intelligence
Machine
Learning
Visualization

Data mining is differ than statistics in kind of data
(not only numerical) , kinds of methods ( mostly use
machine learning methods), more than one
hypotheses, amount of data (statistics uses samples).

Data Mining uses methods from Machine
Learning such as decision tree and neural nets.
 Machine Learning uses samples and Data Mining
uses whole data.
 Data Mining can access data from database.
Machine Learning some times used to replace
human where Data Mining to help human.

Databases part of Data Mining that provide the
fast and reliable access to data.
Databases used for data operation (Storing and
retrieving data), Data Mining for Decision
making.

 Search techniques , Knowledge representation,
Knowledge acquisition, maintenance and
application are other branches of Artificial
Intelligence which are highly related with Data
Mining.

Visualization is used to gain visual insights
into the structure of the data.
 Visualization is in large quantities used as a
pre- and post-processing tool for data mining.

Process of Data Mining
 Data Mining is essentially a process of data
drive extraction of not so obvious but useful
information from large databases.
The entire process is interactive and iterative.

Data Cleaning
Data Integration
Databases
Data Warehouse
Task-relevant Data
Selection
Data Mining
Pattern Evaluation.

Data cleaning
 Real-world data tends to be incomplete, noisy and inconsistent.
 incomplete: lacking attribute values, lacking certain attributes of
interest.
◦ e.g., occupation=“ ” (missing data)
 noisy: containing noise, errors, or outliers
◦ e.g., Salary=“−10” (an error)
 inconsistent: containing difference in codes or names,
◦ e.g., Age=“42” Birthday=“03/07/1997”

Data Integration
 Data integration is the merging of data
from multiple sources.
 These sources may include multiple
databases, data cubes, or flat files.

Data Selection
 Where data relevant to the analysis task are
retrieved from the database. Therefore,
irrelevant, weakly relevant or redundant
attributes may be detected and removed.

Data Transformation
 Where data are transformed or consolidated into forms
appropriate for mining by performing:
 Summary or aggregation operation, for example:
Daily sales may be aggregated to monthly sales or
annual sales.
Generalization, for example:
 City may be generalized to country or age may
generalized to young, middle- age, senior.

Data Mining
 An essential process where intelligent
methods are applied on data to covert it to
knowledge in for decision making.
 Wide range of methods can be used in data
mining such neural nets, decision tree and
Association.

Pattern evaluation
 To identify the truly interesting pattern based on
some interestingness measures.
 A pattern consider interesting if it is:
 Valid
 Novel
Actionable
Understandable

Knowledge Representation
 Knowledge presentation is the framework that
converts a large amount of data into a particular
data or procedure that human being can figure out
based on an intention.
 In Knowledge representation visualization tools
and knowledge representation techniques are used
to present the mined knowledge to the user.

2016
Lecture 3

Data Mining Tasks
 Data mining tasks are the kind of data
patterns that can be mined.
 Data Mining functionalities are used to
specify the kind of patterns to be found in the
data mining tasks.

 In general data mining tasks can be classified into
two categories:
Descriptive mining tasks characterize the general
properties of the data.
Predictive mining tasks perform inferences on the current
data in order to make predictions.
Data Mining Tasks

 Most famous data mining tasks:
 Classification [Predictive]
Prediction [Predictive]
Association Rules [Descriptive]
Clustering [Descriptive]
Outlier Analysis [Descriptive]
Data Mining Tasks

Classification
 Classification is used for predictive mining tasks.
 The input data for predictive modeling consists of
two types of variables:
Explanatory variables, which define the essential properties of
the data.
 Target variables , whose values are to be predicted.
 Classification is used to predicate the value of
discrete target variable.

Prediction
 Similar to classification, except we are trying to predict
the value of a variable (e.g. amount of purchase),
rather than a class (e.g. purchaser or non-purchaser).

Association
 Association Rules aims to find out the relationship
among valuables in database, resulting in deferent types
of rules.
 Seek to produce a set of rules describing the set of
features that are strongly related to each others.

Association
Gender Age Smoker LAD% RCA%
F 52 Y 85 100
M 62 N 80 0
M 75 Y 70 80
M 73 Y 40 99
M 66 N 50 45
… … … … …
 LAD%－ The percentage of heat disease caused by left anterior descending coronary artery.
 RCA%－ The percentage of heat disease caused by right coronary artery.
Original data from a research on heart disease

Association
Medical Association Rules
NO. Rule
1 Gender=M∩Age≥70∩Smoker=YRCA%≥50(40%,100%)
2 Gender=F∩Age＜70∩Smoker=YLAD%≥70(20%,100%)
 Rule 1 indicates：40% of the cases are male, over 70 years old and have the habit of
smoking, the possibility of RCA%≥50% is 100%
 Rule 2 indicates：20% of the cases are female, under 70 years old and have the habit
of smoking, the possibility of LAD%≥70% is 100%

Clustering
 Finds groups of data pointes (clusters) so that data
points that belong to one cluster are more similar to
each other than to data points belonging to different
cluster.

Clustering
Document Clustering:
 Goal: To find groups of documents that are similar to each
other based on the important terms appearing in them.
 Approach: To identify frequently occurring terms in each
document. Form a similarity measure based on the frequencies
of different terms. Use it to cluster.
 Gain: Information Retrieval can utilize the clusters to relate a
new document or search term to clustered documents.

Outlier Analysis
 Discovers data points that are significantly different
than the rest of the data. Such points are known as
anomalies or outliers.

Outline
Data Mining Tasks
Challenges of Data Mining
Data mining application examples
Introduction to RapidMiner

Scalability: Scalable techniques are needed
to handle the massive scale of data.
Dimensionality: Many applications may
involves a large number of dimensions (e.g.
features or attributes of data)

Heterogeneous and Complex Data: In recent years
complicated data types such as graph-based, text-free
and structured data types are introduced. Techniques
developed for data mining must be able to handle the
heterogeneity of the data.

Data Quality: Many data sets are imperfect due to
present of missing values and noise un the data. To
handle the imperfection, robust data mining algorithms
must be developed.

Data Distribution: As the volume of data increases , it
is no longer possible or safe to keep all the data in the
same place. As a result, the need for distributed data
mining techniques has increased over the years.

Privacy Preservation: While privacy intends to prevent
the disclosure of information, data mining attempts to
revel interesting knowledge about data. As a result,
there is growing interest in developing privacy-
preserving data mining algorithms.

Outline
Data Mining Tasks
Data mining application examples
Introduction to RapidMine

Data mining application
Science
astronomy, bioinformatics, drug discovery, …
Business
advertising, CRM (Customer Relationship management),
investments, manufacturing, sports/entertainment, telecom, e-
Commerce, targeted marketing, health care, …
Web
search engines, web mining,…
Government
law enforcement, profiling tax cheaters,

Introduction to-data-mining chapter 1

Introduction to-data-mining chapter 1

Recomendados

Recomendados

Más contenido relacionado

La actualidad más candente

La actualidad más candente (20)

Similar a Introduction to-data-mining chapter 1

Similar a Introduction to-data-mining chapter 1 (20)

Más de Mahmoud Alfarra

Más de Mahmoud Alfarra (20)

Último

Último (20)

Introduction to-data-mining chapter 1