Chapter I Introduction MIS 214 2014 2015 Spring 1

Chapter I: Introduction MIS 214 2014/2015 Spring 1

Chapter 1. Introduction n Motivation: Why data mining? n Methodology of Knowledge Discovery in Databases n Data mining functionalities n Are all the patterns interesting? n Business applications of data mining 2

Motivation: “Necessity is the Mother of Invention” n Data explosion problem n Automated data collection tools and mature database technology lead to tremendous amounts of data stored in databases, data warehouses and other information repositories n Need to convert such data into knowledge and information n Applications n Business management n Production control n Market analysis n Engineering design n Science exploration 3

Evolution of Database Technology (1) n Data collection, database creation n Data management n n n data storage and retrieval database transaction processing Data analysis and understanding n Data mining and data warehousing 4

Evolution of Database Technology (2) n 1960 s: n n 1970 s: n n Data collection, database creation, IMS and network DBMS Relational data model, relational DBMS implementation 1980 s: n n n RDBMS, advanced data models (extended-relational, OO, deductive, etc. ) Application-oriented DBMS (spatial, scientific, engineering, etc. ) 1990 s: n n Data mining, data warehousing, multimedia databases, and Web databases 2000 s n Stream data management and mining n Data mining and its applications n Web technology (XML, data integration) and global information systems 5

n The Explosive Growth of Data: from terabytes to petabytes n Data collection and data availability n n Automated data collection tools, database systems, Web, computerized society Major sources of abundant data n Business: Web, e-commerce, transactions, stocks, … n Science: Remote sensing, bioinformatics, scientific simulation, … n Society and everyone: news, digital cameras, You. Tube n We are drowning in data, but starving for knowledge! n “Necessity is the mother of invention”—Data mining—Automated analysis of massive data sets 6

Developments in computer hardware n Powerful and affordable computers n Data collection equipment n Storage media n Communication and networking 7

Data Warehouse n n n Data cleaning Data integration OLAP: On-Line Analytical Processing n n n summarization consolidation aggregation view information from different angles but additional data analysis tools are needed for n n n classification clustering charecterization of data changing over time 8

Data rich information poor situation n Abundance of data need for powerful data analysis tools “data tombs” - data archives n n Important decisions are made n n seldom visited not on the information rich data stored in databases but on a decision maker’s intuition no tool to extract knowledge embedded in vast amounts of data Expert system technology n n domain experts to input knowledge time consuming and costly 9

What Is Data Mining? n Data mining (knowledge discovery in databases): n n Alternative names and their “inside stories”: n n n Extraction of interesting (non-trivial, implicit, previously unknown and potentially useful) information or patterns from data in large databases Data mining: a misnomer? Knowledge discovery(mining) in databases (KDD), knowledge extraction, data/pattern analysis, data archeology, data dredging, information harvesting, business intelligence, etc. What is not data mining? n n query processing. Expert systems or small ML/statistical programs 10

Data Mining vs. Data Query n Data Query: e. g. n n n A list of all customers who use a credit card to buy a PC A list of all MIS students having a GPA of 3. 5 or higher and has studied 4 or less semesters Data Mining problems: e. g. n n n What is the likelihood of a customer purchasing PC with credit card Given the characteristics of MIS students predict her SPA in the comming term What are the characteristics of MIS undergrad students 11

Chapter 1. Introduction n Motivation: Why data mining? n Methodology of Knowledge Discovery in Databases n Data mining functionalities n Are all the patterns interesting? n Business applications of data mining 12

Why Data Mining? n Four questions to be answered n Can the problem clearly be defined? n Does potentially meaningful data exists? n n Does the data contain hidden knowledge or useful only for reporting purposes? Will the cost of processing the data will be less then the likely increase in profit from the knowledge gained from applying any data mining project 13

Steps of a KDD Process (1) n 1. Goal identification: Define problem n relevant prior knowledge and goals of application 2. Creating a target data set: data selection 3. Data preprocessing: (may take 60%-80% of effort!) n removal of noise or outliers n strategies for handling missing data fields n accounting for time sequence information n n 4. Data reduction and transformation: n Find useful features, dimensionality/variable reduction, invariant representation. 14

Steps of a KDD Process (2) n 5. Data Mining: n n n Choosing functions of data mining: n summarization, classification, regression, association, clustering. Choosing the mining algorithm(s): n which models or parameters Search for patterns of interest 6. Presentation and Evaluation: n visualization, transformation, removing redundant patterns, etc. 7. Taking action: n incorporating into the performance system n documenting n reporting to interested parties 15

An example: Customer Segmentation n n 1. Marketing department wants to perform a segmentation study on the customers of AE Company 2. Decide on revevant variables from a data warehouse on customers, sales, promotions n n n Customers: name, ID, income, age, education, . . . Sales: hisory of sales Promotion: promotion types durations. . . 3. Hendle missing income, addresses. . determine outliers if any 4. Cenerate new index variables representing wealth of customers n n Wealth = a*income+b*#houses+c*#cars. . . Make neccesary transformations z scores so that some data mining algorithms work more efficiently 16

Example: Customer Segmentation cont. n n 5. a: Choose clustering as the data mining functionality as it is the natural one for a segmentation study so as to find group of customers with similar charecteristics 5. b: Choose a clustering algorithm n n 5. c: Apply the algorithm n n n K-means or k-medoids or any suitable one for that problem Find clusters or segments 6. make reverse transformations, visualize the customer segments 7. present the results in the form of a report to the marketing deprtment n n İmplement the segmentation as part of a DSS so that it can be applied repeatedly at certain internvals as new customers arrive Develop marketing strategies for each segment 17

Data Mining: A KDD Process Pattern Evaluation n Data mining: the core Data Mining of knowledge discovery process. Task-relevant Data Warehouse Selection Data Cleaning Data Integration Databases 18

Data Mining in Business Intelligence Increasing potential to support business decisions Decision Making Data Presentation Visualization Techniques End User Business Analyst Data Mining Information Discovery Data Analyst Data Exploration Statistical Summary, Querying, and Reporting Data Preprocessing/Integration, Data Warehouses Data Sources Paper, Files, Web documents, Scientific experiments, Database Systems 15 March 2018 Data Mining: Concepts and Techniques DBA 19 19

Architecture of a Typical Data Mining System Graphical user interface Pattern evaluation Data mining engine Database or data warehouse server Data cleaning & data integration Databases Knowledge-base Filtering Data Warehouse 20

Architecture of a Typical Data Mining System n n n Data base, data warehouse Data base or data warehouse server Knowledge base n n concept hierarchies user beliefs n n n other thresholds Data mining engine n functional modules n n n asses pattern’s interestingness characterization, association, classification, cluster analysis, evolution and deviation analysis Pattern evaluation module Graphical user interface 21

Data Mining: Confluence of Multiple Disciplines Database Technology Machine Learning Information Science Statistics Data Mining Visualization Other Disciplines 22

Why Confluence of Multiple Disciplines? n Tremendous amount of data n n High-dimensionality of data n n Micro-array may have tens of thousands of dimensions High complexity of data n n n n Algorithms must be highly scalable to handle such as tera-bytes of data Data streams and sensor data Time-series data, temporal data, sequence data Structure data, graphs, social networks and multi-linked data Heterogeneous databases and legacy databases Spatial, spatiotemporal, multimedia, text and Web data Software programs, scientific simulations New and sophisticated applications 15 March 2018 Data Mining: Concepts and Techniques 23 23

Efficient and Scalable Techniques n n n For an algorithm to be efficient and scalable its running time should be predictable and acceptable How n n Parallel and distributed algorithms Sampling from databases 24

Chapter 1. Introduction n Motivation: Why data mining? n Methodology of Knowledge Discovery in Databases n Data mining functionalities n Are all the patterns interesting? n Business applications of data mining 25

Two Styles of Data Mining n Descriptive data mining n n Predictive data mining n n n perform inference on the current data to make predictions we know what to predict Not mutually exclusive n n characterize the general properties of the data in the database finds patterns in data and the user determines which ones are important used together Descriptive predictive Eg. Customer segmentation – descriptive by clustering Followed by a risk assignment model – predictive by ANN 26

Supervised vs. Unsupervised Learning n n Supervised learning (classification, prediction) n Supervision: The training data (observations, measurements, etc. ) are accompanied by labels indicating the class of the observations n New data is classified based on the training set Unsupervised learning (summarization. association, clustering) n The class labels of training data is unknown n Given a set of measurements, observations, etc. with the aim of establishing the existence of classes or clusters in the data 27

Descriptive Data Mining (1) n n n Discovering new patterns inside the data Used during the data exploration steps Typical questions answered by descriptive data mining n n n what is in the data what does it look like are there any unusual patterns what dose the data suggest for customer segmentation users may have no idea n which kind of patterns may be interesting 28

Descriptive Data Mining (2) n patterns at verious granularities n geograph n n student n n country - city - region - street university - faculty - department - minor Fuctionalities of descriptive data mining n Clustering n n Ex: customer segmentation summarization visualization Association n Ex: market basket analysis 29

A model is a black box X: vector of independent variables or inputs Y =f(X) : an unknown function Y: dependent variables or output a single variable or a vector inputs X 1, X 2 Model Y output The user does not care what the model is doing it is a black box interested in the accuracy of its predictions 30

Predictive Data Mining (1) n Using known examples the model is trained n n the more data with known outcomes is available n n n the unknown function is learned from data the better the predictive power of the model Used to predict outcomes whose inputs are known but the output values are not realized yet Never %100 accurate 31

Predictive Data Mining (2) n The performance of a model on past data is not important n n to predict the known outcomes Its performance on unknown data is much more important 32

Typical questions answered by predictive models n Who is likely to respond to our next offer n n n Which customers are likely to leave in the next six months What transactions are likely to be fraudulent n n based on history of previous marketing campaigns based on known examples of fraud What is the total amount spending of a customer in the next month 33

Data Mining Functionalities (1) n Concept description: Characterization and discrimination n n Generalize, summarize, and contrast data characteristics, e. g. , big spenders vs. budget spenders Association (correlation and causality) n n n Multi-dimensional vs. single-dimensional association age(X, “ 20. . 29”) ^ income(X, “ 20. . 29 K”) àbuys(X, “PC”) [support = 2%, confidence = 60%] contains(T, “computer”) àcontains(x, “software”) [1%, 75%] 34

Data Mining Functionalities (2) n Classification and Numerical-Prediction n n Finding models (functions) that describe and distinguish classes or concepts for future prediction E. g. , classify people as healty or sick, or classify transactions as fraudulent or not n n n Methods: decision-tree, classification rule, neural network Prediction: Predict some unknown or missing numerical values Cluster analysis n n Class label is unknown: Group data to form new classes, e. g. , cluster customers of a retail company to learn about characteristics of different segments Clustering based on the principle: maximizing the intra-class similarity and minimizing the interclass similarity 35

Data Mining Functionalities (3) n Outlier analysis n Outlier: a data object that does not comply with the general behavior of the data n It can be considered as noise or exception but is quite useful in fraud detection, rare events analysis n Trend and evolution analysis n n Sequential pattern mining: click stream analysis n n Trend and deviation: regression analysis Similarity-based analysis Other pattern-directed or statistical analyses 36

Concept Description n Characterization Discerimination Data n n n classes of items for sale n n classes or concpets computers, printers concepts of customers: n n big. Spenders Budget. Spenders 37

Data Characterization n n Summarization the data of the class under study (target class) Methods n n SQL queries OLAP roll up -operation n user-controlled data summarization along a specified dimension n without step by step user interraction n attribute oriented induction the output of characterization n n pie charts, bar chars, curves, multidimensional data cube, or cross tabs in rule form as characteristic rules 38

Characterization example n Description summarizing the characteristics of customers who spend more than $1000 a year at All. Elecronics n n age, employment, income drill down on any dimension n on occupation view these according to their type of employment 39

Data Discrimination n Comparing the target class with one or a set of comparative classes (contrasting classes) n n n these classes can be specified by the use database queries methods and output n n similar to those used for characterization include comparative measures to distinguish between the target and contrasting classes 40

Discrimination examples n Example 1: Compare the general features of software products n n n whose sales increased by %10 in the last year (target class) whose sales decreased by at least %30 during the same period (contrasting class) Example 2: Compare two groups of AE customers n I) who shop for computer products regularly (target class) n n II) who rarely shop for such products (contrasting class) n n n less than three times a year The resulting description: %80 of I group customers n n n more than two times a month university education ages 20 -40 %60 of II group customers n n seniors or young no university degree 41

Multidimensional Data sales according to region month and product type Dimensions: Product, Location, Time Hierarchical summarization paths gi on n Re Industry Region Year Product Category Country Quarter Product City Office Month Week Day Month 42

Association Analysis n n n Discovery of association rules showing attribute-value conditions that occur frequently together in a given set of data widely used n market basket n transaction data analysis more formally X Y that is A 1 A 2. . Ak B 1 B 2. . Bl A 1 , B 1 are attribute value pairs or predicates 43

Example: association analysis n n n From the All. Es database n age(X, ” 20. . 29”) income(X, ” 1, 000. . . 2, 000”) buy(X, ”CD player”) n (support = %2, n confidence= %60) X is a variable representing a customer %2 of the AE customers are n between 20 and 29 age n incomes ranging from 1 to 2 billon TL n buy CD player with %60 probability that customers in those age and income groups will buy CD player a multidimensional association rule n contains more than one attribute or predicate 44

Market basket analysis n n customers buying behaviour is investigated Based on only the transactions data n n no information about customer properties: age income Managers n are interested in which products or product groups are sold together 45

Transactional Database Transaction ID Item List 10001 Computer, CD, pritner 10002 Ploter, monitor, mouse 10003 Computer, DVD Player 10004 Printer 10005 Ploter, UPS, modem 46

Example: basket analysis rule n n n buy(computer) buy(printer) (support= %1, confidence=%60) %1 of all transactions contains n n if a transaction contains computer n n contains a single predicate an association rule is interesting if n n n there is a %60 chance that it contains printer as well a single dimensional association rule n n computer and printer its support exceeds a minimum threshold and its confidence exceeds a min threshold These min values are set by specialists 47

Classification n Learning is supervised Dependent variable is categorical Build a model able to assign new instances to one of a set of well-defined classes 48

Typical Classification Problems n n n Given characteristics of individuals differentiate them who have suffered a heart attack from those who have not Determine if a credit card purchase is fraudulent Classify a car loan applicant as a good or a poor credit risk 49

Methods of Classification n Decision Trees Artificial Neural Networks Bayesian Classification n n Naïve Belief Networks k-nearest neighbor Regression n Logistic (logit) probit n n Predicts probability of each class when the dependent variable is categorical n good customer bed customer or employed unemployed 50

Steps of classification process n (1) Train the model n n n (2) Test the model n n n using a training set data objects whose class labels are known on a test sample whose class labels are known but not used for training the model (3) Use the model for classification new data whose class labels are unknown 51

An example - classification ID age income education Historical data Each customer type İs known Each customer has a Label Type 1 35 800 udergrad risky 2 26 600 High. Sch risky 3 48 1200 grad normal 8 52 2500 udergrad good 44 29 1700 High. Sch good Cust. ID age income education Type 17 43 550 Ph. D. risky 27 68 1650 grad Normal Cust. ID age income 11 36 850 27 28 1650 Educatin Type Udergrd ? grad ? Testing set whose labels are also n. Known but not used in model n. Training the model n New customers Whose type hsa to be n. Estimated n. Each new customer hss to be classified as Risky normal or good n 52

Orginal data 53

Historical data Each customer type İs known Each customer has a Label Testing set whose labels are also n. Known but not used in model n. Training the model n New customers Whose type hsa to be n. Estimated n. Each new customer hss to be classified as buyer or non buyer n 54

An example – classification cont. n Based on historical data develop a classification model n n Decision tree, neural network, regression. . . Test the performance of the model on a portion of the historical data İf accuricy of the model is satisfactory Use the model on the new customers n 11 and 27 to assign a type these new customers 55

Example AE customers age goodl risky Yearly income 56

Example AE customers age goodl risky ? Assign the new customer whose type in unknown to either * or + Yearly income 57

Solution x 2 : age good risky 35 x 1 : yearly income 1000 rule: IF yearly income> 1000 and age> 35 THEN good ELSE risky 58

Credit Card Promotion Policy n Credit card companies n n n Promotional offerings with their monthly credit card billing Offers provide the opportunity to purchase items such as magazines, … A data mining study n n Predict individual behaviour What is the likelihood of an individual towards taking the advantage of promotions based on individual characteristics, credit history. . Expected reduction in postage; paper and processing costs for the credit card company 59

Credit Card Promotion Database Income Range Magazıne Promotıon Watch Promotıon Lıfe Insurance Promotıon Gender Age Credıt Card Insurance 40 -50 K Yes No No Male 45 No 30 -40 K Yes Yes Female 40 No 40 -50 K No No No Male 42 No 30 -40 K Yes Yes Male 43 Yes 50 -60 K Yes No Yes Female 38 No 20 -30 K No No No Female 55 No 30 -40 K Yes No Yes Male 35 Yes 20 -30 K No Yes No Male 27 No 30 -40 K Yes No No Male 43 No 30 -40 K Yes Yes Female 41 No 40 -50 K No Yes Female 43 No 20 -30 K No Yes Male 29 No 50 -60 K Yes Yes Female 39 No 40 -50 K No Yes No Male 55 No 20 -30 K No No Yes Female 19 Yes 60

Decision Trees for Credit Card Insurance Database age <=43 Dependent Variable Life Insurance Promotion >43 Gender Female N 0, Y 6 Decision: Yes N 3, Y 0 Decision: No Male A Production Rule from the Tree Cr Ins No N 4, Y 1 Decision: No critical value of 43 nis deter by the nalgorithm n Yes IF (age<=43)&(Sex=Male) &(Credit Card In = No) THEN Life Insurance Pr = No Yes 2, No 0 Decision? Yes 61

Artificial Neural Networks n n Set of interconnected nodes designed to imitate the functioning of the human brain Feed-forward network n Supervised learner model 62

For the promotion example n n Encode all variables Assign a numerical value even for qualitative variables such as sex Say X 1 represent gender When n n Male X 1 =1 Female X 1 =0 63

Input layer X 1=+1 1 Hidden layer Output layer W 1, 5=0. 014 5 W 5, 9=-0. 17 X 2=0 X 3=0. 5 X 4=-1 (1 -0. 78)2 is error square 1 actual value of O 9 for a particular Data object 0. 78 is predicted value 64

Weights updating n n Weights between nodes are adjusted so as to reduce error Details of the training process for neural networks are not important for the time being 65

Numerical-Prediction n n Similar to classification Output is a continuous variable Estimation: current value Prediction: future outcome rather then current behavior 66

Typical Numerical Prediction Problems n n n Estimate the salary of an individual who owns a sports car Predict next week`s closing price for the IMKB 100 index Forecast next days temperature 67

Numerical Prediction methods n n Artificial Neural networks linear regression n n non-linear regression n n Yi = a 0+a 1 X 1, i+a 2 X 2, i+. . . +ak. Xk, i+ui Yi =f(X 1, i, X 2, i, . . , Xk, ia 1, a 2, . . , ak, ui) generalized linear regression n logistic n n poisson regression n n logit, probit for count variables Regression Trees 68

Example: Prediction and Classification n Classification is used to classify customers applying for credit cards n n known class labels: risky, reliable when a new customer applies looking at her charecteristics n income age education wealth region. . . n n Customer class is predicted n independent variables Prediction: The monthly expense of a new customer ( a real continuous variable ) is predicted based on personal information n n income education wealth profession. . . Some are numeric some categorical 69

Cluster Analysis n Class label is unknown: Group data to form new classes, n assign class labels to each data object n n n e. g. , cluster customers to find customer segments Clustering based on the principle: maximizing the intra-class similarity and minimizing the interclass similarity n n n Unknown generated by the clustering model Objects within a cluster have high similarity in comparison to one another but are very dissimilar to objects in other clusters there may be hierarchy of classes 70

Example: Clustering n n n Can be performed on AE customer data to identify homogenous subpopulations of customers represent individual target groups for marketing 71

Before clustering After clustering 72

distance Type 1 Type 2 type 3 income Clustering according to income and distance to store three cluster of data points are evident 73

Outlier Analysis n Outlier: a data object that does not comply with the general behavior of the data n It can be considered as noise or exception but is quite useful in fraud detection, rare events analysis n DECTECED using n n distance measures n n statistical tests visually inspecting the data Examples: 74

Reasons for outliers n n Measurement errors coding errors n n age is entered as 999 nature of data n n salary of the general manager is much more higher than the other employees in crisis the interest rate was in the order of 1000 s 75

Evolution Analysis n Describes and models regularities or trends for objects whose behavior changes over time n Distinct features include n n Sequential pattern mining, periodicity analysis n n Trend and deviation: time-series data analysis Similarity-based analysis Example n Stock market predictions: future stock prices n for overall stocks: indexes or individual company stocks 76

Sequential Pattern Analysis n n n Determine sequential patterns in data Based on time sequence of actions Similar to associations n n Relationship is based on time Example 1: buy CD player today buy CD within one week Example 2: In what sequence web pages of an ebusiness company are accessed %70 percents of visitors follows n n A B C or A D B C or A E B C He then determines to add a link directly from page A to page C 77

Chapter 1. Introduction n Motivation: Why data mining? n Methodology of Knowledge Discovery in Databases n Data mining functionalities n Are all the patterns interesting? n Business applications of data mining 78

Are All the “Discovered” Patterns Interesting? n A data mining system/query may generate thousands of patterns, not all of them are interesting. n Are all patterns interesting? n Typically not -only a small fraction of patterns are interesting to any given user n Interestingness measures: A pattern is interesting if n it is easily understood by humans, n valid on new or test data with some degree of certainty, n potentially useful, n novel, or n validates some hypothesis that a user seeks to confirm 79

Objective vs. subjective interestingness measures: n n Objective: based on statistics and structures of patterns, e. g. , n support, n X Y P(X Y): probability of a transaction contains both X and Y n confidence, degree of certainty of the detected association n P(Y I X) the conditional probability : the probability that a transaction containing X also contains Y n thresholds - controlled by the user n ex: rules that do not satisfy a confidence threshold of %50 are uninteresting Subjective: based on user’s belief in the data, e. g. , unexpectedness, novelty, actionability, etc. 80

Chapter 1. Introduction n Motivation: Why data mining? n Methodology of Knowledge Discovery in Databases n Data mining functionalities n Are all the patterns interesting? n Business Applications of data mining 81

Potential Business Applications n Market analysis and management n n target marketing, customer relation management, market basket analysis, cross selling, market segmentation Risk analysis and management n Banks assume a financial risk when they grant loans n n n risk models attempt to predict the probability of default or fail to pay back the borrowed amount Credit cards Insurance companies n Fraud detection and management n Other Applications n Text mining (news group, email, documents) and Web analysis. n Intelligent query answering 82

Market Analysis and Management (1) n Where are the data sources for analysis? n n Credit card transactions, loyalty cards, discount coupons, customer complaint calls, plus (public) lifestyle studies, clickstreams Customer profiling-segmentation n data mining can tell you what types of customers buy what products (clustering or classification) n Target marketing n Find clusters of “model” customers who share the same characteristics: interest, income level, spending habits, etc. 83

Market Analysis and Management (2) n Effectiveness of sales campaigns n n Advertisements, coupons, discounts, bonuses promote products and attract customers can help improve profits Compare amount of sales and number of transactions n n during the sales period versus before or after the sales campaign Association analysis n which items are likely to be purchased together with the items on sale 84

Market Analysis and Management (3) n Customer retention Analysis of Customer loyalty n n n sequences of purchases of particular customers goods purchased at different periods by the same customers can be grouped into sequences changes in customer consumption or loyalty suggests adjustments on the pricing and variety of goods to retain old customers and attract new customers Cross-selling and up-selling n n n associations from sales records a customer who buy a PC is likely to buy a printer purchase recommendations 85

Fraud Detection and Management n Applications n n Approach n n widely used in health care, retail, credit card services, telecommunications (phone card fraud), etc. use historical data to build models of fraudulent behavior and use data mining to help identify similar instances Examples n n n Credit card transactions: The FALCON fraud assessment system by HNC Inc. to signal possibly fraudulent credit card transactions money laundering: detect suspicious money transactions (US Treasury's Financial Crimes Enforcement Network) Detecting telephone fraud: ASPECT European Research Gr. n n Unsupervised clustering to detect fraud in mobile phone networks Telephone call model: destination of the call, duration, time of day or week. Analyze patterns that deviate from an expected norm. 86

Health Care n Storing patients` records in electronic format, developments in medical information systems n n Regularities, trends and surprising events extracted by data mining methods n n n Large amount of clinical data ANN, temporal reasoning assist clinicians to make informed decisions and improving health sevices MERCK-MEDCO Managed Care, Pharmaceutical Insurance … company n Uncover less expensive but equally effective drug treatments 87

Financial Data Analysis n n Financial data n complete, reliable, high quality Loan payment prediction and customer credit policy analysis 88

Loan payment prediction and customer credit policy analysis n Factors influencing loan payment performance n n n n n loan-to-value ratio term of the loan dept ratio (total monthly debt/total monthly income) payment-to-income ratio income level education level residence region credit history analysis may find that n n payment-income ratio is a dominant factor while education level and debt ratio are not 89

Risk Management and Insurance n n determine insurance rates manage investment portfolios differentiate between companies and/or individuals who are good and poor credit risks Farmer`s Group discover a scenario: n n Someone who owns a sports car is not a higher accident risk Conditions: the sport car to be a second car and the family car to be a station wagon or a sedan 90

Data Mining for the Telecommunication Industry n Telecommunication data are multidimensional n n n n duration location of callee used to identify and compare n n calling-time location of caller type of call data traffic resource usage profit system workload user group behavior fraudulent pattern analysis and identification of unusual patterns to achieve customer loyalty characteristics of customers affecting line usage 91

Other Applications n Sports and Gaming n n Text Mining n n Predicting outcome of football games Spam detection Internet Web Mining n Web usage mining n n n İmprove link structure Recommander Systmes Web structure mining: mining link structure of Web 92

Other Applications n Educational Data Mining n n n Clustering students Design enterece exams, selection policies Human Resources n n How to select applicants Online Dating n Recommandataions to visitors 93

Summary n n n Data mining: discovering interesting patterns from large amounts of data A natural evolution of database technology, in great demand, with wide applications A KDD process includes data cleaning, data integration, data selection, transformation, data mining, pattern evaluation, and knowledge presentation Mining can be performed in a variety of information repositories Data mining functionalities: characterization, discrimination, association, classification, clustering, 94