Data Mining Data Lecture Notes for Chapter 2
![Data Mining: Data Lecture Notes for Chapter 2 Introduction to Data Mining by Tan, Data Mining: Data Lecture Notes for Chapter 2 Introduction to Data Mining by Tan,](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-1.jpg)
![What is Data? l Collection of data objects and their attributes l An attribute What is Data? l Collection of data objects and their attributes l An attribute](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-2.jpg)
![Attribute Values l Attribute values are numbers or symbols assigned to an attribute l Attribute Values l Attribute values are numbers or symbols assigned to an attribute l](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-3.jpg)
![Types of Attributes l There are different types of attributes – Nominal u Examples: Types of Attributes l There are different types of attributes – Nominal u Examples:](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-4.jpg)
![Attribute Level Transformation Nominal Any permutation of values If all employee ID numbers were Attribute Level Transformation Nominal Any permutation of values If all employee ID numbers were](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-5.jpg)
![Discrete and Continuous Attributes l Discrete Attribute – Has only a finite or countably Discrete and Continuous Attributes l Discrete Attribute – Has only a finite or countably](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-6.jpg)
![Types of data sets l l l Record – Data Matrix – Distance Matrix Types of data sets l l l Record – Data Matrix – Distance Matrix](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-7.jpg)
![Record Data l Data that consists of a collection of records, each of which Record Data l Data that consists of a collection of records, each of which](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-8.jpg)
![Data Matrix l If data objects have the same fixed set of numeric attributes, Data Matrix l If data objects have the same fixed set of numeric attributes,](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-9.jpg)
![Data Matrix Example for Documents l Each document becomes a `term' vector, – each Data Matrix Example for Documents l Each document becomes a `term' vector, – each](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-10.jpg)
![Distance Matrix © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 11 Distance Matrix © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 11](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-11.jpg)
![Transaction Data l A special type of record data, where – each record (transaction) Transaction Data l A special type of record data, where – each record (transaction)](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-12.jpg)
![Graph Data l Examples: Generic graph – how can we create this? (hint: think Graph Data l Examples: Generic graph – how can we create this? (hint: think](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-13.jpg)
![Data Quality What kinds of data quality problems? l How can we detect problems Data Quality What kinds of data quality problems? l How can we detect problems](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-14.jpg)
![Noise l Noise refers to modification of original values – Examples: distortion of a Noise l Noise refers to modification of original values – Examples: distortion of a](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-15.jpg)
![Outliers l Outliers are data objects with characteristics that are considerably different than most Outliers l Outliers are data objects with characteristics that are considerably different than most](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-16.jpg)
![Missing Values l Reasons for missing values – Information is not collected (e. g. Missing Values l Reasons for missing values – Information is not collected (e. g.](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-17.jpg)
![Duplicate Data l Data set may include data objects that are duplicates, or almost Duplicate Data l Data set may include data objects that are duplicates, or almost](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-18.jpg)
![Data Preprocessing Standardization or Normalization l Aggregation l Sampling l Dimensionality Reduction l Feature Data Preprocessing Standardization or Normalization l Aggregation l Sampling l Dimensionality Reduction l Feature](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-19.jpg)
![Data Transformations targeted at minimizing experimental variance l Normalization: scaled to fall within a Data Transformations targeted at minimizing experimental variance l Normalization: scaled to fall within a](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-20.jpg)
![Data Transformation: Normalization l min-max normalization l z-score normalization l normalization by decimal scaling Data Transformation: Normalization l min-max normalization l z-score normalization l normalization by decimal scaling](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-21.jpg)
![Aggregation l Combining two or more attributes (or objects) into a single attribute (or Aggregation l Combining two or more attributes (or objects) into a single attribute (or](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-22.jpg)
![Sampling l Sampling is the main technique employed for data selection. – It is Sampling l Sampling is the main technique employed for data selection. – It is](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-23.jpg)
![Sampling … l The key principle for effective sampling is the following: – using Sampling … l The key principle for effective sampling is the following: – using](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-24.jpg)
![Types of Sampling l Simple Random Sampling – There is an equal probability of Types of Sampling l Simple Random Sampling – There is an equal probability of](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-25.jpg)
![Sample Size 8000 points © Tan, Steinbach, Kumar 2000 Points Introduction to Data Mining Sample Size 8000 points © Tan, Steinbach, Kumar 2000 Points Introduction to Data Mining](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-26.jpg)
![Sample Size l What sample size is necessary to get at least one object Sample Size l What sample size is necessary to get at least one object](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-27.jpg)
![Dimensionality Reduction l Why? – When dimensionality increases, data becomes increasingly sparse in the Dimensionality Reduction l Why? – When dimensionality increases, data becomes increasingly sparse in the](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-28.jpg)
![Principal Components Analysis: Intuition Goal is to find a projection that captures the largest Principal Components Analysis: Intuition Goal is to find a projection that captures the largest](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-29.jpg)
![Principal Component Analysis: Details l Let A be an n ×n matrix representing the Principal Component Analysis: Details l Let A be an n ×n matrix representing the](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-30.jpg)
![Feature Subset Selection l Another way to reduce dimensionality of data l Redundant features Feature Subset Selection l Another way to reduce dimensionality of data l Redundant features](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-31.jpg)
![Simple Discretization Methods: Binning l Equal-width (distance) partitioning: – Divides the range into N Simple Discretization Methods: Binning l Equal-width (distance) partitioning: – Divides the range into N](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-32.jpg)
![Information/Entropy l Given probabilitites p 1, p 2, . . , ps whose sum Information/Entropy l Given probabilitites p 1, p 2, . . , ps whose sum](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-33.jpg)
![Entropy-Based Discretization l Given a set of samples S, if S is partitioned into Entropy-Based Discretization l Given a set of samples S, if S is partitioned into](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-34.jpg)
![Distance Metrics Revisited: Minkowski Distance l Minkowski Distance is a generalization of Euclidean Distance Distance Metrics Revisited: Minkowski Distance l Minkowski Distance is a generalization of Euclidean Distance](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-35.jpg)
![Mahalanobis Distance is the covariance matrix of the input data X For red points, Mahalanobis Distance is the covariance matrix of the input data X For red points,](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-36.jpg)
![Mahalanobis Distance Example (work this out yourselves) Covariance Matrix: C A: (0. 5, 0. Mahalanobis Distance Example (work this out yourselves) Covariance Matrix: C A: (0. 5, 0.](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-37.jpg)
![Common Properties of a Distance l Distances, such as the Euclidean distance, have some Common Properties of a Distance l Distances, such as the Euclidean distance, have some](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-38.jpg)
![Common Properties of a Similarity l Similarities, also have some well known properties. 1. Common Properties of a Similarity l Similarities, also have some well known properties. 1.](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-39.jpg)
![Density l Density-based clustering require a notion of density l Examples: – Euclidean density Density l Density-based clustering require a notion of density l Examples: – Euclidean density](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-40.jpg)
![Euclidean Density – Cell-based l Simplest approach is to divide region into a number Euclidean Density – Cell-based l Simplest approach is to divide region into a number](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-41.jpg)
![Euclidean Density – Center-based l Euclidean density is the number of points within a Euclidean Density – Center-based l Euclidean density is the number of points within a](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-42.jpg)
- Slides: 42
![Data Mining Data Lecture Notes for Chapter 2 Introduction to Data Mining by Tan Data Mining: Data Lecture Notes for Chapter 2 Introduction to Data Mining by Tan,](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-1.jpg)
Data Mining: Data Lecture Notes for Chapter 2 Introduction to Data Mining by Tan, Steinbach, Kumar (edited by S. Parthasarathy 3/2007) © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 1
![What is Data l Collection of data objects and their attributes l An attribute What is Data? l Collection of data objects and their attributes l An attribute](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-2.jpg)
What is Data? l Collection of data objects and their attributes l An attribute is a property or characteristic of an object Attributes – Examples: eye color of a person, temperature, etc. – Attribute is also known as variable, field, characteristic, or feature Objects l A collection of attributes describe an object or a record – Object is also known as record, point, case, sample, entity, or instance © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 2
![Attribute Values l Attribute values are numbers or symbols assigned to an attribute l Attribute Values l Attribute values are numbers or symbols assigned to an attribute l](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-3.jpg)
Attribute Values l Attribute values are numbers or symbols assigned to an attribute l Distinction between attributes and attribute values – Same attribute can be mapped to different attribute values u Example: height can be measured in feet or meters – Different attributes can be mapped to the same set of values Example: Attribute values for ID and age are integers u But properties of attribute values can be different u – ID has no limit but age has a maximum and minimum value © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 3
![Types of Attributes l There are different types of attributes Nominal u Examples Types of Attributes l There are different types of attributes – Nominal u Examples:](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-4.jpg)
Types of Attributes l There are different types of attributes – Nominal u Examples: ID numbers, eye color, zip codes – Ordinal u Examples: rankings (e. g. , taste of potato chips on a scale from 1 -10), grades, height in {tall, medium, short} – Interval u Examples: calendar dates – Ratio u Examples: length, time, counts © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 4
![Attribute Level Transformation Nominal Any permutation of values If all employee ID numbers were Attribute Level Transformation Nominal Any permutation of values If all employee ID numbers were](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-5.jpg)
Attribute Level Transformation Nominal Any permutation of values If all employee ID numbers were reassigned, would it make any difference? Ordinal An order preserving change of values, i. e. , new_value = f(old_value) where f is a monotonic function. An attribute encompassing the notion of good, better best can be represented equally well by the values {1, 2, 3} or by { 0. 5, 1, 10}. Interval new_value =a * old_value + b where a and b are constants Calendar dates can be converted – financial vs. gregorian etc. new_value = a * old_value Length can be measured in meters or feet. Ratio Comments
![Discrete and Continuous Attributes l Discrete Attribute Has only a finite or countably Discrete and Continuous Attributes l Discrete Attribute – Has only a finite or countably](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-6.jpg)
Discrete and Continuous Attributes l Discrete Attribute – Has only a finite or countably infinite set of values – Examples: zip codes, counts, or the set of words in a collection of documents – Often represented as integer variables. – Note: binary attributes are a special case of discrete attributes l Continuous Attribute – Has real numbers as attribute values – Examples: temperature, height, or weight. – Practically, real values can only be measured and represented using a finite number of digits. – Continuous attributes are typically represented as floating-point variables. © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 6
![Types of data sets l l l Record Data Matrix Distance Matrix Types of data sets l l l Record – Data Matrix – Distance Matrix](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-7.jpg)
Types of data sets l l l Record – Data Matrix – Distance Matrix – Document Data – Transaction Data Graph – World Wide Web – Molecular Structures Ordered – Spatial Data – Temporal Data – Sequential Data – Genetic Sequence Data © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 7
![Record Data l Data that consists of a collection of records each of which Record Data l Data that consists of a collection of records, each of which](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-8.jpg)
Record Data l Data that consists of a collection of records, each of which consists of a fixed set of attributes © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 8
![Data Matrix l If data objects have the same fixed set of numeric attributes Data Matrix l If data objects have the same fixed set of numeric attributes,](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-9.jpg)
Data Matrix l If data objects have the same fixed set of numeric attributes, then the data objects can be thought of as points in a multi-dimensional space, where each dimension represents a distinct attribute l Such data set can be represented by an m by n matrix, where there are m rows, one for each object, and n columns, one for each attribute © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 9
![Data Matrix Example for Documents l Each document becomes a term vector each Data Matrix Example for Documents l Each document becomes a `term' vector, – each](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-10.jpg)
Data Matrix Example for Documents l Each document becomes a `term' vector, – each term is a component (attribute) of the vector, – the value of each component is the number of times the corresponding term occurs in the document. © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 10
![Distance Matrix Tan Steinbach Kumar Introduction to Data Mining 4182004 11 Distance Matrix © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 11](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-11.jpg)
Distance Matrix © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 11
![Transaction Data l A special type of record data where each record transaction Transaction Data l A special type of record data, where – each record (transaction)](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-12.jpg)
Transaction Data l A special type of record data, where – each record (transaction) involves a set of items. – For example, consider a grocery store. The set of products purchased by a customer during one shopping trip constitute a transaction, while the individual products that were purchased are the items. © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 12
![Graph Data l Examples Generic graph how can we create this hint think Graph Data l Examples: Generic graph – how can we create this? (hint: think](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-13.jpg)
Graph Data l Examples: Generic graph – how can we create this? (hint: think about a distance matrix) 2 1 5 2 5 © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 13
![Data Quality What kinds of data quality problems l How can we detect problems Data Quality What kinds of data quality problems? l How can we detect problems](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-14.jpg)
Data Quality What kinds of data quality problems? l How can we detect problems with the data? l What can we do about these problems? l l Examples of data quality problems: – Noise and outliers – missing values – duplicate data © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 14
![Noise l Noise refers to modification of original values Examples distortion of a Noise l Noise refers to modification of original values – Examples: distortion of a](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-15.jpg)
Noise l Noise refers to modification of original values – Examples: distortion of a person’s voice when talking on a poor phone and “snow” on television screen Two Sine Waves © Tan, Steinbach, Kumar Two Sine Waves + Noise Introduction to Data Mining Frequency Plot (FFT) 4/18/2004 15
![Outliers l Outliers are data objects with characteristics that are considerably different than most Outliers l Outliers are data objects with characteristics that are considerably different than most](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-16.jpg)
Outliers l Outliers are data objects with characteristics that are considerably different than most of the other data objects in the data set © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 16
![Missing Values l Reasons for missing values Information is not collected e g Missing Values l Reasons for missing values – Information is not collected (e. g.](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-17.jpg)
Missing Values l Reasons for missing values – Information is not collected (e. g. , people decline to give their age and weight) – Attributes may not be applicable to all cases (e. g. , annual income is not applicable to children) l Handling missing values – – Eliminate Data Objects Estimate Missing Values Ignore the Missing Value During Analysis Replace with all possible values (weighted by their probabilities) © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 17
![Duplicate Data l Data set may include data objects that are duplicates or almost Duplicate Data l Data set may include data objects that are duplicates, or almost](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-18.jpg)
Duplicate Data l Data set may include data objects that are duplicates, or almost duplicates of one another – Major issue when merging data from heterogeous sources l Examples: – Same person with multiple email addresses l Data cleaning – Process of dealing with duplicate data issues © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 18
![Data Preprocessing Standardization or Normalization l Aggregation l Sampling l Dimensionality Reduction l Feature Data Preprocessing Standardization or Normalization l Aggregation l Sampling l Dimensionality Reduction l Feature](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-19.jpg)
Data Preprocessing Standardization or Normalization l Aggregation l Sampling l Dimensionality Reduction l Feature subset selection l Discretization and Binarization l Attribute Transformation l © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 19
![Data Transformations targeted at minimizing experimental variance l Normalization scaled to fall within a Data Transformations targeted at minimizing experimental variance l Normalization: scaled to fall within a](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-20.jpg)
Data Transformations targeted at minimizing experimental variance l Normalization: scaled to fall within a small, specified range – min-max normalization – z-score normalization – normalization by decimal scaling l Centralization: – Based on fitting a distribution to the data – Distance function between distributions u KL Distance u Mean Centering © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 20
![Data Transformation Normalization l minmax normalization l zscore normalization l normalization by decimal scaling Data Transformation: Normalization l min-max normalization l z-score normalization l normalization by decimal scaling](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-21.jpg)
Data Transformation: Normalization l min-max normalization l z-score normalization l normalization by decimal scaling Where j is the smallest integer such that Max(| © Tan, Steinbach, Kumar Introduction to Data Mining |)<1 4/18/2004 21
![Aggregation l Combining two or more attributes or objects into a single attribute or Aggregation l Combining two or more attributes (or objects) into a single attribute (or](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-22.jpg)
Aggregation l Combining two or more attributes (or objects) into a single attribute (or object) l Purpose – Data reduction u Reduce the number of attributes or objects – Change of scale u Cities aggregated into regions, states, countries, etc – More “stable” data u Aggregated data tends to have less variability © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 22
![Sampling l Sampling is the main technique employed for data selection It is Sampling l Sampling is the main technique employed for data selection. – It is](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-23.jpg)
Sampling l Sampling is the main technique employed for data selection. – It is often used for both the preliminary investigation of the data and the final data analysis. l Statisticians sample because obtaining the entire set of data of interest is too expensive or time consuming. l Sampling is used in data mining because processing the entire set of data of interest is too expensive or time consuming. © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 23
![Sampling l The key principle for effective sampling is the following using Sampling … l The key principle for effective sampling is the following: – using](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-24.jpg)
Sampling … l The key principle for effective sampling is the following: – using a sample will work almost as well as using the entire data sets, if the sample is representative – A sample is representative if it has approximately the same property (of interest) as the original set of data © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 24
![Types of Sampling l Simple Random Sampling There is an equal probability of Types of Sampling l Simple Random Sampling – There is an equal probability of](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-25.jpg)
Types of Sampling l Simple Random Sampling – There is an equal probability of selecting any particular item l Sampling without replacement – As each item is selected, it is removed from the population l Sampling with replacement – Objects are not removed from the population as they are selected for the sample. In sampling with replacement, the same object can be picked up more than once u l Stratified sampling – Split the data into several partitions; then draw random samples from each partition © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 25
![Sample Size 8000 points Tan Steinbach Kumar 2000 Points Introduction to Data Mining Sample Size 8000 points © Tan, Steinbach, Kumar 2000 Points Introduction to Data Mining](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-26.jpg)
Sample Size 8000 points © Tan, Steinbach, Kumar 2000 Points Introduction to Data Mining 500 Points 4/18/2004 26
![Sample Size l What sample size is necessary to get at least one object Sample Size l What sample size is necessary to get at least one object](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-27.jpg)
Sample Size l What sample size is necessary to get at least one object from each of 10 groups. l Learning Curve © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 27
![Dimensionality Reduction l Why When dimensionality increases data becomes increasingly sparse in the Dimensionality Reduction l Why? – When dimensionality increases, data becomes increasingly sparse in the](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-28.jpg)
Dimensionality Reduction l Why? – When dimensionality increases, data becomes increasingly sparse in the space that it occupies – Definitions of density and distance between points, which is critical for clustering and outlier detection, become less meaningful l Objectives: – – l Avoid curse of dimensionality Reduce amount of time and memory required by data mining algorithms Allow data to be more easily visualized May help to eliminate irrelevant features or reduce noise Techniques – Principle Component Analysis or Singular Value Decomposition – Others: supervised and non-linear techniques © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 28
![Principal Components Analysis Intuition Goal is to find a projection that captures the largest Principal Components Analysis: Intuition Goal is to find a projection that captures the largest](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-29.jpg)
Principal Components Analysis: Intuition Goal is to find a projection that captures the largest amount of variation in data l Find the eigenvectors of the covariance matrix l The eigenvectors define the new space l x 2 e x 1 © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 29
![Principal Component Analysis Details l Let A be an n n matrix representing the Principal Component Analysis: Details l Let A be an n ×n matrix representing the](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-30.jpg)
Principal Component Analysis: Details l Let A be an n ×n matrix representing the correlation or covariance of the data. – λ is an eigenvalue of A if there exists a non-zero vector v such that: Av = λ v often rewritten as (A- λI)v=0 l In this case, vector v is called an eigenvector of A corresponding to λ. For each eigenvalue λ, the set of all vectors v satisfying Av = λ v is called the eigenspace of A corresponding to λ. © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 30
![Feature Subset Selection l Another way to reduce dimensionality of data l Redundant features Feature Subset Selection l Another way to reduce dimensionality of data l Redundant features](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-31.jpg)
Feature Subset Selection l Another way to reduce dimensionality of data l Redundant features – duplicate much or all of the information contained in one or more other attributes – Example: purchase price of a product and the amount of sales tax paid l Irrelevant features – contain no information that is useful for the data mining task at hand – Example: students' ID is often irrelevant to the task of predicting students' GPA © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 31
![Simple Discretization Methods Binning l Equalwidth distance partitioning Divides the range into N Simple Discretization Methods: Binning l Equal-width (distance) partitioning: – Divides the range into N](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-32.jpg)
Simple Discretization Methods: Binning l Equal-width (distance) partitioning: – Divides the range into N intervals of equal size: uniform grid – if A and B are the lowest and highest values of the attribute, the width of intervals will be: W = (B –A)/N. – The most straightforward, but outliers may dominate presentation – Skewed data is not handled well. l Equal-depth (frequency) partitioning: – Divides the range into N intervals, each containing approximately same number of samples – Good data scaling – Managing categorical attributes can be tricky. © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 32
![InformationEntropy l Given probabilitites p 1 p 2 ps whose sum Information/Entropy l Given probabilitites p 1, p 2, . . , ps whose sum](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-33.jpg)
Information/Entropy l Given probabilitites p 1, p 2, . . , ps whose sum is 1, Entropy is defined as: l Entropy measures the amount of randomness or surprise or uncertainty. Only takes into account non-zero probabilities l © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 33
![EntropyBased Discretization l Given a set of samples S if S is partitioned into Entropy-Based Discretization l Given a set of samples S, if S is partitioned into](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-34.jpg)
Entropy-Based Discretization l Given a set of samples S, if S is partitioned into two intervals S 1 and S 2 using boundary T, the entropy after partitioning is l The boundary that minimizes the entropy function over all possible boundaries is selected as a binary discretization. The process is recursively applied to partitions obtained until some stopping criterion is met, e. g. , l l Experiments show that it may reduce data size and improve classification accuracy © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 34
![Distance Metrics Revisited Minkowski Distance l Minkowski Distance is a generalization of Euclidean Distance Distance Metrics Revisited: Minkowski Distance l Minkowski Distance is a generalization of Euclidean Distance](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-35.jpg)
Distance Metrics Revisited: Minkowski Distance l Minkowski Distance is a generalization of Euclidean Distance Where r is a parameter, n is the number of dimensions (attributes) and pk and qk are, respectively, the kth attributes (components) or data objects p and q. l Eucledian and Manhattan are specializations of this metric (r=2 or 1 respectively) © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 35
![Mahalanobis Distance is the covariance matrix of the input data X For red points Mahalanobis Distance is the covariance matrix of the input data X For red points,](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-36.jpg)
Mahalanobis Distance is the covariance matrix of the input data X For red points, the Euclidean distance is 14. 7, Mahalanobis distance is 6. © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 36
![Mahalanobis Distance Example work this out yourselves Covariance Matrix C A 0 5 0 Mahalanobis Distance Example (work this out yourselves) Covariance Matrix: C A: (0. 5, 0.](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-37.jpg)
Mahalanobis Distance Example (work this out yourselves) Covariance Matrix: C A: (0. 5, 0. 5) B B: (0, 1) A C: (1. 5, 1. 5) Mahal(A, B) = 5 Mahal(A, C) = 4 © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 37
![Common Properties of a Distance l Distances such as the Euclidean distance have some Common Properties of a Distance l Distances, such as the Euclidean distance, have some](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-38.jpg)
Common Properties of a Distance l Distances, such as the Euclidean distance, have some well known properties. 1. d(p, q) 0 for all p and q and d(p, q) = 0 only if p = q. (Positive definiteness) 2. 3. d(p, q) = d(q, p) for all p and q. (Symmetry) d(p, r) d(p, q) + d(q, r) for all points p, q, and r. (Triangle Inequality) where d(p, q) is the distance (dissimilarity) between points (data objects), p and q. l A distance that satisfies these properties is a metric © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 38
![Common Properties of a Similarity l Similarities also have some well known properties 1 Common Properties of a Similarity l Similarities, also have some well known properties. 1.](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-39.jpg)
Common Properties of a Similarity l Similarities, also have some well known properties. 1. s(p, q) = 1 (or maximum similarity) only if p = q. 2. s(p, q) = s(q, p) for all p and q. (Symmetry) where s(p, q) is the similarity between points (data objects), p and q. © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 39
![Density l Densitybased clustering require a notion of density l Examples Euclidean density Density l Density-based clustering require a notion of density l Examples: – Euclidean density](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-40.jpg)
Density l Density-based clustering require a notion of density l Examples: – Euclidean density u Euclidean density = number of points per unit volume – Probability density – Graph-based density © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 40
![Euclidean Density Cellbased l Simplest approach is to divide region into a number Euclidean Density – Cell-based l Simplest approach is to divide region into a number](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-41.jpg)
Euclidean Density – Cell-based l Simplest approach is to divide region into a number of rectangular cells of equal volume and define density as # of points the cell contains © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 41
![Euclidean Density Centerbased l Euclidean density is the number of points within a Euclidean Density – Center-based l Euclidean density is the number of points within a](https://slidetodoc.com/presentation_image/bc10c36122d8a8798bc7d4546ce4efde/image-42.jpg)
Euclidean Density – Center-based l Euclidean density is the number of points within a specified radius of the point © Tan, Steinbach, Kumar Introduction to Data Mining 4/18/2004 42
Bayesian classification in data mining lecture notes
Data mining lecture notes
Data mining lecture notes
Data mining lecture notes
01:640:244 lecture notes - lecture 15: plat, idah, farad
Mining complex types of data in data mining
Mining multimedia databases in data mining
Exploratory data analysis lecture notes
Strip mining vs open pit mining
Mineral resources and mining chapter 13
Difference between strip mining and open pit mining
Text and web mining
Project procurement management lecture notes
Theology proper lecture notes
Public sector accounting definition
4 p's of software project management
Electricity and magnetism lecture notes
Physics 111 lecture notes
Physical science lecture notes
Power system dynamics and stability lecture notes
Microbial physiology lecture notes
Application of mechatronics ppt
Limits fits and tolerances
Money-time relationship and equivalence
Bjt transistor notes
Requirement analysis in software engineering notes
Introduction to ofdm
Land use planning '' lecture notes
Project management lecture notes doc
Lecture notes on homiletics
Foundation engineering lecture notes
Image processing lecture notes
Intermediate microeconomics lecture notes
Cloud computing lecture
Decision theory lecture notes
Polynomial regression least squares
Advanced inorganic chemistry lecture notes
50f7 tolerance
Global stiffness matrix
Goodrich method flood routing
Adf.test in r
Shape memory alloys lecture notes
Notes on research methods