© Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy...
-
Upload
juniper-nichols -
Category
Documents
-
view
216 -
download
0
Transcript of © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy...
![Page 1: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/1.jpg)
© Prentice Hall 1
CIS 674Introduction to Data Mining
Srinivasan [email protected]
Office Hours: TTH 2-3:18PM DL317
![Page 2: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/2.jpg)
© Prentice Hall 2
Introduction Outline
• Define data mining
• Data mining vs. databases
• Basic data mining tasks
• Data mining development
• Data mining issues
Goal:Goal: Provide an overview of data mining. Provide an overview of data mining.
![Page 3: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/3.jpg)
© Prentice Hall 3
Introduction
• Data is produced at a phenomenal rate• Our ability to store has grown• Users expect more sophisticated information• How?
UNCOVER HIDDEN INFORMATIONUNCOVER HIDDEN INFORMATION
DATA MININGDATA MINING
![Page 4: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/4.jpg)
© Prentice Hall 4
Data Mining
• Objective: Fit data to a model
• Potential Result: Higher-level meta information that may not be obvious when looking at raw data
• Similar terms– Exploratory data analysis– Data driven discovery– Deductive learning
![Page 5: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/5.jpg)
© Prentice Hall 5
Data Mining Algorithm
• Objective: Fit Data to a Model– Descriptive– Predictive
• Preferential Questions – Which technique to choose?
• ARM/Classification/Clustering• Answer: Depends on what you want to do with data?
– Search Strategy – Technique to search the data• Interface? Query Language?• Efficiency
![Page 6: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/6.jpg)
© Prentice Hall 6
Database Processing vs. Data Mining Processing
• Query– Well defined– SQL
• Query– Poorly defined– No precise query language
OutputOutput– PrecisePrecise– Subset of databaseSubset of database
OutputOutput– FuzzyFuzzy– Not a subset of databaseNot a subset of database
![Page 7: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/7.jpg)
© Prentice Hall 7
Query Examples
• Database
• Data Mining
– Find all customers who have purchased milkFind all customers who have purchased milk
– Find all items which are frequently purchased Find all items which are frequently purchased with milk. (association rules)with milk. (association rules)
– Find all credit applicants with last name of Smith.Find all credit applicants with last name of Smith.– Identify customers who have purchased more Identify customers who have purchased more than $10,000 in the last month.than $10,000 in the last month.
– Find all credit applicants who are poor credit Find all credit applicants who are poor credit risks. (classification)risks. (classification)– Identify customers with similar buying habits. Identify customers with similar buying habits. (Clustering)(Clustering)
![Page 8: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/8.jpg)
© Prentice Hall 8
Data Mining Models and Tasks
![Page 9: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/9.jpg)
© Prentice Hall 9
Basic Data Mining Tasks• Classification maps data into predefined groups
or classes– Supervised learning– Pattern recognition– Prediction
• Regression is used to map a data item to a real valued prediction variable.
• Clustering groups similar data together into clusters.– Unsupervised learning– Segmentation– Partitioning
![Page 10: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/10.jpg)
© Prentice Hall 10
Basic Data Mining Tasks (cont’d)
• Summarization maps data into subsets with associated simple descriptions.– Characterization– Generalization
• Link Analysis uncovers relationships among data.– Affinity Analysis– Association Rules– Sequential Analysis determines sequential patterns.
![Page 11: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/11.jpg)
© Prentice Hall 11
Ex: Time Series Analysis• Example: Stock Market• Predict future values• Determine similar patterns over time• Classify behavior
![Page 12: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/12.jpg)
© Prentice Hall 12
Data Mining vs. KDD
• Knowledge Discovery in Databases (KDD): process of finding useful information and patterns in data.
• Data Mining: Use of algorithms to extract the information and patterns derived by the KDD process.
![Page 13: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/13.jpg)
Knowledge Discovery Process
– Data mining: the core of knowledge discovery process.
Data Cleaning
Data Integration
Databases
Preprocessed Data
Task-relevant DataData transformations
Selection
Data Mining
Knowledge Interpretation
![Page 14: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/14.jpg)
© Prentice Hall 14
KDD Process Ex: Web Log• Selection:
– Select log data (dates and locations) to use
• Preprocessing: – Remove identifying URLs– Remove error logs
• Transformation: – Sessionize (sort and group)
• Data Mining: – Identify and count patterns– Construct data structure
• Interpretation/Evaluation:– Identify and display frequently accessed sequences.
• Potential User Applications:– Cache prediction– Personalization
![Page 15: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/15.jpg)
© Prentice Hall 15
Data Mining Development•Similarity Measures•Hierarchical Clustering•IR Systems•Imprecise Queries•Textual Data•Web Search Engines
•Bayes Theorem•Regression Analysis•EM Algorithm•K-Means Clustering•Time Series Analysis
•Neural Networks•Decision Tree Algorithms
•Algorithm Design Techniques•Algorithm Analysis•Data Structures
•Relational Data Model•SQL•Association Rule Algorithms•Data Warehousing•Scalability Techniques
HIGH PERFORMANCE
DATA MINING
![Page 16: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/16.jpg)
© Prentice Hall 16
KDD Issues
• Human Interaction
• Overfitting
• Outliers
• Interpretation
• Visualization
• Large Datasets
• High Dimensionality
![Page 17: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/17.jpg)
© Prentice Hall 17
KDD Issues (cont’d)
• Multimedia Data
• Missing Data
• Irrelevant Data
• Noisy Data
• Changing Data
• Integration
• Application
![Page 18: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/18.jpg)
© Prentice Hall 18
Social Implications of DM
• Privacy
• Profiling
• Unauthorized use
![Page 19: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/19.jpg)
© Prentice Hall 19
Data Mining Metrics
• Usefulness
• Return on Investment (ROI)
• Accuracy
• Space/Time
![Page 20: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/20.jpg)
© Prentice Hall 20
Database Perspective on Data Mining
• Scalability
• Real World Data
• Updates
• Ease of Use
![Page 21: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/21.jpg)
© Prentice Hall 21
Outline of Today’s Class
• Statistical Basics– Point Estimation– Models Based on Summarization– Bayes Theorem– Hypothesis Testing– Regression and Correlation
• Similarity Measures
![Page 22: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/22.jpg)
© Prentice Hall 22
Point Estimation• Point Estimate: estimate a population
parameter.• May be made by calculating the parameter for a
sample.• May be used to predict value for missing data.• Ex:
– R contains 100 employees– 99 have salary information– Mean salary of these is $50,000– Use $50,000 as value of remaining employee’s
salary. Is this a good idea?
![Page 23: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/23.jpg)
© Prentice Hall 23
Estimation Error
• Bias: Difference between expected value and actual value.
• Mean Squared Error (MSE): expected value of the squared difference between the estimate and the actual value:
• Why square?• Root Mean Square Error (RMSE)
![Page 24: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/24.jpg)
© Prentice Hall 24
Jackknife Estimate• Jackknife Estimate: estimate of parameter is
obtained by omitting one value from the set of observed values.– Treat the data like a population– Take samples from this population– Use these samples to estimate the parameter
• Let θ(hat) be an estimate on the entire pop.• Let θ(j)(hat) be an estimator of the same form
with observation j deleted• Allows you to examine the impact of outliers!
![Page 25: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/25.jpg)
© Prentice Hall 25
Maximum Likelihood Estimate (MLE)
• Obtain parameter estimates that maximize the probability that the sample data occurs for the specific model.
• Joint probability for observing the sample data by multiplying the individual probabilities. Likelihood function:
• Maximize L.
![Page 26: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/26.jpg)
© Prentice Hall 26
MLE Example
• Coin toss five times: {H,H,H,H,T}
• Assuming a perfect coin with H and T equally
likely, the likelihood of this sequence is:
• However if the probability of a H is 0.8 then:
![Page 27: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/27.jpg)
© Prentice Hall 27
MLE Example (cont’d)• General likelihood formula:
• Estimate for p is then 4/5 = 0.8
![Page 28: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/28.jpg)
© Prentice Hall 28
Expectation-Maximization (EM)
• Solves estimation with incomplete data.
• Obtain initial estimates for parameters.
• Iteratively use estimates for missing data and continue until convergence.
![Page 29: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/29.jpg)
© Prentice Hall 29
EM Example
![Page 30: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/30.jpg)
© Prentice Hall 30
EM Algorithm
![Page 31: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/31.jpg)
© Prentice Hall 31
Bayes Theorem Example• Credit authorizations (hypotheses):
h1=authorize purchase, h2 = authorize after further identification, h3=do not authorize, h4= do not authorize but contact police
• Assign twelve data values for all combinations of credit and income:
• From training data: P(h1) = 60%; P(h2)=20%;
P(h3)=10%; P(h4)=10%.
1 2 3 4 Excellent x1 x2 x3 x4 Good x5 x6 x7 x8 Bad x9 x10 x11 x12
![Page 32: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/32.jpg)
© Prentice Hall 32
Bayes Example(cont’d)• Training Data:
ID Income Credit Class xi 1 4 Excellent h1 x4 2 3 Good h1 x7 3 2 Excellent h1 x2 4 3 Good h1 x7 5 4 Good h1 x8 6 2 Excellent h1 x2 7 3 Bad h2 x11 8 2 Bad h2 x10 9 3 Bad h3 x11 10 1 Bad h4 x9
![Page 33: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/33.jpg)
© Prentice Hall 33
Bayes Example(cont’d)• Calculate P(xi|hj) and P(xi)
• Ex: P(x7|h1)=2/6; P(x4|h1)=1/6; P(x2|h1)=2/6; P(x8|h1)=1/6; P(xi|h1)=0 for all other xi.
• Predict the class for x4:– Calculate P(hj|x4) for all hj. – Place x4 in class with largest value.– Ex:
• P(h1|x4)=(P(x4|h1)(P(h1))/P(x4) =(1/6)(0.6)/0.1=1.
• x4 in class h1.
![Page 34: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/34.jpg)
© Prentice Hall 34
Other Statistical Measures
• Chi-Squared– O – observed value– E – Expected value based on hypothesis.
• Jackknife Estimate– estimate of parameter is obtained by omitting one value from the
set of observed values.
• Regression– Predict future values based on past values– Linear Regression assumes linear relationship exists.
y = c0 + c1 x1 + … + cn xn• Find values to best fit the data
• Correlation
![Page 35: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/35.jpg)
© Prentice Hall 35
Similarity Measures
• Determine similarity between two objects.• Similarity characteristics:
• Alternatively, distance measure measure how unlike or dissimilar objects are.
![Page 36: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/36.jpg)
© Prentice Hall 36
Similarity Measures
![Page 37: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/37.jpg)
© Prentice Hall 37
Distance Measures
• Measure dissimilarity between objects
![Page 38: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/38.jpg)
© Prentice Hall 38
Information Retrieval
• Information Retrieval (IR): retrieving desired information from textual data.
• Library Science• Digital Libraries• Web Search Engines• Traditionally keyword based• Sample query:
Find all documents about “data mining”.
DM: Similarity measures; Mine text/Web data.
![Page 39: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/39.jpg)
© Prentice Hall 39
Information Retrieval (cont’d)
• Similarity: measure of how close a query is to a document.
• Documents which are “close enough” are retrieved.
• Metrics:– Precision = |Relevant and Retrieved|
|Retrieved|– Recall = |Relevant and Retrieved|
|Relevant|
![Page 40: © Prentice Hall1 CIS 674 Introduction to Data Mining Srinivasan Parthasarathy srini@cse.ohio-state.edu Office Hours: TTH 2-3:18PM DL317.](https://reader035.fdocuments.in/reader035/viewer/2022062516/56649d945503460f94a7c160/html5/thumbnails/40.jpg)
© Prentice Hall 40
IR Query Result Measures and Classification
IR Classification