Data Mining : Implementations

1. Data Mining : Implementations

2. Constructing Classifier for B.I. The leading Credit Card Company Visa International and Computer Company Acer Inc. has agreed to market the newly launched Laptop Computer. • For targeted marketing campaign, both the companies would like to use the existing database of Visa Credit card holders, to find out most likely buyers of their Computer. • For a given Database, your job will be to construct a Classifier. • This Classifier should take an input as the customer-details like • Age, • Income • Student or Professional • Credit_Card Ratings • And based on above inputs, the Classifier would Predict, whether the person is likely buyer of the Laptop Computer or not.

3. Identifying appropriate technique in D.M. • To construct suitable Classifier • Given an Application, to construct classifier domain . • Top Ten Algorithm in Data Mining were published in 2008 with related research issues (Journal of Knowledge and Information System) : Best Impact Algorithms !. • Classification & Prediction • Classifier with high tolerance for missing values • High dimensionality of the data • Lazy Learners Vs. Eager Learners • Classifying accuracy • Increasing accuracy: bagging & boosting • Various Classifiers • Algorithms • B.I. -Case Study • Finding other Classifiers

4. Top Ten International Data Mining researchers have identified 10 most influential Algorithms for Association Analysis, Classification , Clustering, Statistical Learning, Link Mining; • C4.5 & DTC (Decision Tree Classification ) • K Means Algorithm • Support Vector Machines • Apriori Algorithm • EM ( Expectation Maximization Algorithm) • PageRank Algorithm • AdaBoost : Ensemble Learning approach • K Nearest Neighborhood Algorithm • Naïve Bayes • CART (Classification and Regression Tree)

5. Classification vs. Prediction • Classification: • classifies data (constructs a model) based on the training set and the values (class labels) in a classifying attribute and uses it in classifying new data. • predicts categorical class labels. • Prediction:-Prediction can be viewed as the construction and use of a model to assess the class of an unlabeled sample, or to assess the value or value ranges of an attribute that a given sample is likely to have. • CLASSIFICATION & REGRESSION are two prediction methods. ( discrete Vs. Continuous) • models continuous-valued functions, i.e., predicts unknown or missing values.

6. Examples of Classification Task • Classifying whether the new credit card applicant will be of High, Medium or Low credit risk. • Will be a perspective buyer ? • For a targeted campaign . • Classifying credit card transactions as legitimate or fraudulent. • Categorizing news stories as finance, weather, entertainment, sports, etc • Predicting tumor cells as benign or malignant • Classifying secondary structures of protein as alpha-helix, beta-sheet, or random coil

7. Data Classification — A Two-Step Process • Model construction: describing a set of predetermined classes [ Learning ] • Each tuple / sample is assumed to belong to a predefined class, as determined by the class label attribute. • The set of tuples used for model construction: training set (given data). • The model is represented as classification rules, decision trees, or mathematical formulae. • Model usage: for classifying future or unknown objects [ Classification ] • Estimate accuracy of the model • The known label of test sample is compared with the classified result from the model • Accuracy rate is the percentage of test set samples that are correctly classified by the model • Test set is independent of training set, otherwise over-fitting will occur

8. Training Data Classifier (Model) Classification Process (1): Model Construction ( Learning) Classification Algorithms Classification rules IF rank = ‘professor’ OR years > 6 THEN tenured = ‘yes’

9. Classifier Test Data Unseen Data Classification Process (2): Use the Model in Prediction ( classification) (Jeff, Professor, 4) Tenured?

10. Constructing Classifier for B.I. The leading Credit Card Company Visa International and Computer Company Acer Inc. has agreed to market the newly launched Laptop Computer. • For targeted marketing campaign, both the companies would like to use the existing database of Visa Credit card holders, to find out most likely buyers of their Computer. • For a given Database, your job will be to construct a Classifier. • This Classifier should take an input as the customer-details like • Age, • Income • Student or Professional • Credit_Card Ratings • And based on above inputs, the Classifier would Predict, whether the person is likely buyer of the Laptop Computer or not.

11. To construct Classifier from a given training data

12. Constructing Classifier • Data Domain Analysis • Small Dataset with 04 discrete valued attributes. • The Class label attributes are Boolean in nature! • Class Labels: • C1:buys_computer = ‘yes’ • C2:buys_computer = ‘no’ • Predictive mechanism needed ! • Business application with transactional database: no missing values ! • Implementation details • Execution of Java program with dataset. • Constructing IF…. THEN… ELSE Rules.

13. Data Mining: Concepts and Techniques What is classification? What is prediction? Issues regarding classification and prediction Classification by decision tree induction Bayesian classification Rule-based classification Soft Computing Approaches: Classification by back propagation Support Vector Machines (SVM) Associative classification Lazy learners (or learning from your neighbors) Other classification methods Prediction Accuracy and error measures Ensemble methods Model selection Summary Classification and Prediction

14. Data Mining: Concepts and Techniques Classification in Large Databases • Classification—a classical problem extensively studied by statisticians and machine learning researchers • Scalability: Classifying data sets with millions of examples and hundreds of attributes with reasonable speed • Why decision tree induction in data mining? • relatively faster learning speed (than other classification methods) • convertible to simple and easy to understand classification rules • can use SQL queries for accessing databases • comparable classification accuracy with other methods

15. DT induction: Issues • There are 04 attributes in the table, which attribute to start with at Root ? • Attribute Selection measure or measure of goodness of split • Select the attribute with the highest information gain • Calculateinformation gain of each of 04 attributes by calculating; • Expected information (entropy) • Information needed • Information Gained= (a)-(b) • Find out the maximum I.G. out of attributes age, income, student, credit_rating

16. Data Mining: Concepts and Techniques Class P: buys_computer = “yes” Class N: buys_computer = “no” means “age <=30” has 5 out of 14 samples, with 2 yes’es and 3 no’s. Hence Similarly, Looking at other Gains, we can see the Gain(age) is max. Attribute Selection: Information Gain

17. Gain Ratio for Attribute Selection (C4.5) • Information gain measure is biased towards attributes with a large number of values • C4.5 (a successor of ID3) uses gain ratio to overcome the problem (normalization to information gain) • GainRatio(A) = Gain(A)/SplitInfo(A) • Ex. • gain_ratio(income) = 0.029/0.926 = 0.031 • The attribute with the maximum gain ratio is selected as the splitting attribute

18. Data Mining: Concepts and Techniques age? <=30 overcast >40 31..40 Output: A Decision Tree for “Buys_Computer”

19. Data Mining: Concepts and Techniques age? <=30 overcast >40 31..40 student? credit rating? yes excellent fair no yes no yes no yes Output: A Decision Tree for “buys_computer”

21. Data Mining: Concepts and Techniques Issues: Evaluating Classification Methods • Accuracy • classifier accuracy: predicting class label • predictor accuracy: guessing value of predicted attributes • Speed • time to construct the model (training time) • time to use the model (classification/prediction time) • Robustness: handling noise and missing values • Scalability: efficiency in disk-resident databases • Interpretability • understanding and insight provided by the model • Other measures, e.g., goodness of rules, such as decision tree size or compactness of classification rules

22. Constructing Classifier for B.I. The leading Credit Card Company Visa International and Computer Company Acer Inc. has agreed to market the newly launched Laptop Computer. • For targeted marketing campaign, both the companies would like to use the existing database of Visa Credit card holders, to find out most likely buyers of their Computer. • For a given Database, your job will be to construct a Classifier. • This Classifier should take an input as the customer-details like • Age, • Income • Student or Professional • Credit_Card Ratings • And based on above inputs, the Classifier would Predict, whether the person is likely buyer of the Laptop Computer or not. • DTC was proposed as one of the alternative, but it showed some limitations.

23. Credit Card Database & BI application • As the Visa International has accumulated huge record of its Credit card holders for last 05 years amounting to terabytes of data. • Though larger the data better the training of the Classifiers ( some popularly used DTC algorithms C4.5 uses as much as 2/3 of the data for training the Classifiers). • In Case of our BI application, the problem starts with calculating Information Gain • Every attributes involves all transactions for IG • disk resident training data is not fitting into our memory for training activity. • Scalability Problem!

24. DTC • Speed was not good enough ! • Because of swapping of training data in and out of memory the time to construct the model (training time) ! • time to use the model (prediction time) was although reasonable ! • The B.I. Solution for Customer having Age age <=30, Income = medium, Student = yes and Credit_rating = Fair) falling into either of the category with level of certainty was not specifically available.

25. Training Dataset Class Labels: C1:buys_computer = ‘yes’ C2:buys_computer = ‘no’ Data sample: X = (age <=30, Income = medium, Student = yes Credit_rating = Fair) Belongs to which Class?

26. Data Mining: Concepts and Techniques What is classification? What is prediction? Issues regarding classification and prediction Classification by decision tree induction Bayesian classification Bayesian Belief Network Soft Computing techniques Classification by back propagation Support Vector Machines (SVM) Associative classification Lazy learners (or learning from your neighbors) Other classification methods Prediction Accuracy and error measures Ensemble methods Model selection Summary Classification : Which one now ?

27. Domain Analysis for Credit Card Database • Predictive mechanism needed .. • No missing values... • Attributes may or may not depend on each other • Database is huge ! • Attributes are NOT many !! • Class Labels are Boolean !!! • Which Classifier? And please Why?

28. Using Bayesian Classification for B.I. • Bayesian classifiers are statistical classifiers which predict class membership probabilities such as the probability that a given sample belongs to a particular class. • Bayesian classification is based on the Bayes theorem and it is observed that a simple Bayesian Classifier known as the naïve Bayesian classifier to be comparable in performance with decision tree and Neural network classifiers. • Naïve Bayesian classifier assume that the effect of an attribute value on a given class is independent of the values of the other attributes. (conditional independence). .

29. Data Mining: Concepts and Techniques Bayesian Theorem: Basics • Let X be a data sample (“evidence”): class label is unknown • Let H be a hypothesisthat X belongs to class C • Classification is to determine P(H|X), the probability that the hypothesis holds given the observed data sample X • P(H) (prior probability), the initial probability • E.g., Xwill buy computer, regardless of age, income, … • P(X): probability that sample data is observed 2) P(X|H) (posteriori probability), the probability of observing the sample X, given that the hypothesis holds ( X belongs to Class C) • E.g., Given that X will buy computer, the prob. that X is 31..40, medium income

30. Data Mining: Concepts and Techniques Bayesian Theorem • Given training dataX, posteriori probability of a hypothesis H, P(H|X), follows the Bayes theorem • Informally, this can be written as posteriori = likelihood x prior/evidence • Predicts Xbelongs to C2iff the probability P(Ci|X) is the highest among all the P(Ck|X) for all the k classes • Practical difficulty: require initial knowledge of many probabilities, significant computational cost

31. Towards Naïve Bayesian Classifier • Let D be a training set of tuples and their associated class labels, and each tuple is represented by an n-D attribute vector X = (x1, x2, …, xn) Suppose there are m classes C1, C2, …, Cm.we have C1,C2 • Classification is to derive the maximum posteriori, i.e., the maximal P(Ci|X) • This can be derived from Bayes’ theorem • Since P(X) is constant for all classes, only needs to be maximized

32. Derivation of Naïve Bayes Classifier • A simplified assumption: attributes are conditionally independent • This greatly reduces the computation cost: Only counts the class distribution • If Ak is categorical, P(xk|Ci) is the # of tuples in Ci having value xk for Ak divided by |Ci, D| (# of tuples of Ci in D) • If Ak is continous-valued, P(xk|Ci) is usually computed based on Gaussian distribution with a mean μ and standard deviation σ and P(xk|Ci) is

33. Data Mining: Concepts and Techniques Naïve Bayesian Classifier: Training Dataset Class Labels: C1:buys_computer = ‘yes’ C2:buys_computer = ‘no’ Data sample X = (age <=30, Income = medium, Student = yes Credit_rating = Fair)

34. Data Mining: Concepts and Techniques Naïve Bayesian Classifier: An Example • P(Ci): P(buys_computer = “yes”) = 9/14 = 0.643 P(buys_computer = “no”) = 5/14= 0.357 • Compute P(X|Ci) for each class P(age = “<=30” | buys_computer = “yes”) = 2/9 = 0.222 P(age = “<= 30” | buys_computer = “no”) = 3/5 = 0.6 P(income = “medium” | buys_computer = “yes”) = 4/9 = 0.444 P(income = “medium” | buys_computer = “no”) = 2/5 = 0.4 P(student = “yes” | buys_computer = “yes) = 6/9 = 0.667 P(student = “yes” | buys_computer = “no”) = 1/5 = 0.2 P(credit_rating = “fair” | buys_computer = “yes”) = 6/9 = 0.667 P(credit_rating = “fair” | buys_computer = “no”) = 2/5 = 0.4 • X = (age <= 30 , income = medium, student = yes, credit_rating = fair) P(X|Ci) : P(X|buys_computer = “yes”) = 0.222 x 0.444 x 0.667 x 0.667 = 0.044 P(X|buys_computer = “no”) = 0.6 x 0.4 x 0.2 x 0.4 = 0.019 P(X|Ci)*P(Ci) : P(X|buys_computer = “yes”) * P(buys_computer = “yes”) = 0.028 P(X|buys_computer = “no”) * P(buys_computer = “no”) = 0.007 Therefore, X belongs to class (“buys_computer = yes”)

35. Customer Prediction: Problems • There may be field data having some records in which probability of buying a computer by a Low income may come out to be Zero for all other attributes ! • The Bayes Theorem assumes that the attribute-values like age, income,credit_ratings etc are conditionally independent ! • Above assumption may not hold true if the additional attributes are added in due course of time, some of which have inter-dependencies / associations.

36. Avoiding the 0-Probability Problem • Suppose a dataset with 1000 tuples, income=low (0), income= medium (990), and income = high (10), • Naïve Bayesian prediction requires each conditional prob. be non-zero. Otherwise, the predicted prob. will be zero • How to avoid this 0-Probability Problem !!

37. Laplacian Probability • Use Laplacian correction (or Laplacian estimator) • Adding 1 to each case Prob(income = low) = 1/1003 Prob(income = medium) = 991/1003 Prob(income = high) = 11/1003 • This incorrect/manipulated data may incur in erroneous result , but still will be usable ! • The “corrected” prob. estimates are close to their “uncorrected” counterparts

38. Data Mining: Concepts and Techniques Naïve Bayesian Classifier: Comments • Advantages • Easy to implement • Good results obtained in most of the cases • Disadvantages • Assumption: class conditional independence, therefore loss of accuracy • Practically, dependencies exist among variables • E.g., hospitals: patients: Profile: age, family history, etc. Symptoms: fever, cough etc., Disease: lung cancer, diabetes, etc. • Dependencies among these cannot be modeled by Naïve Bayesian Classifier • How to deal with these dependencies? • Bayesian Belief Networks

39. Data Mining: Concepts and Techniques Y Z P Bayesian Belief Networks • Bayesian belief network allows a subset of the variables conditionally independent • A graphical model of causal relationships • Represents dependency among the variables • Gives a specification of joint probability distribution • Nodes: random variables • Links: dependency • X and Y are the parents of Z, and Y is the parent of P • No dependency between Z and P • Has no loops or cycles X

40. Data Mining: Concepts and Techniques Scalable Decision Tree Induction Methods • SLIQ (EDBT’96 — Mehta et al.) • Builds an index for each attribute and only class list and the current attribute list reside in memory • SPRINT (VLDB’96 — J. Shafer et al.) • Constructs an attribute list data structure • PUBLIC (VLDB’98 — Rastogi & Shim) • Integrates tree splitting and tree pruning: stop growing the tree earlier • RainForest (VLDB’98 — Gehrke, Ramakrishnan & Ganti) • Builds an AVC-list (attribute, value, class label) • BOAT (PODS’99 — Gehrke, Ganti, Ramakrishnan & Loh) • Uses bootstrapping to create several small samples

41. Decision Tree Classifiers : Survey paper

42. Data Mining: Concepts and Techniques What is classification? What is prediction? Issues regarding classification and prediction Classification by decision tree induction Bayesian classification Support Vector Machines (SVM) Rule-based classification Classification by back propagation Associative classification Lazy learners (or learning from your neighbors) Other classification methods Prediction Accuracy and error measures Ensemble methods Model selection Summary Classification and Prediction

43. Bayesian Classification • Bayesian classifiers are statistical classifiers which predict class membership probabilities such as the probability that a given sample belongs to a particular class. • Bayesian classification is based on the Bayes theorem and it is observed that a simple Bayesian Classifier known as the naïve Bayesian classifier to be comparable in performance with decision tree and Neural network classifiers. • Naïve Bayesian classifier assume that the effect of an attribute value on a given class is independent of the values of the other attributes. (conditional independence). • Bayesian belief networks are graphical models, which unlike naïve bayesian classifiers allow the representation of dependencies among subsets of attributes. Can be Used for classification.

44. Bayesian Belief Networks (II) • Bayesian belief network allows a subset of the variables conditionally independent • A graphical model of causal relationships • Several cases of learning Bayesian belief networks • Given both network structure and all the variables: easy • Given network structure but only some variables • When the network structure is not known in advance

45. Data Mining: Concepts and Techniques Y Z P Bayesian Belief Networks • Bayesian belief network allows a subset of the variables conditionally independent • A graphical model of causal relationships • Represents dependency among the variables • Gives a specification of joint probability distribution • Nodes: random variables • Links: dependency • X and Y are the parents of Z, and Y is the parent of P • No dependency between Z and P • Has no loops or cycles X

46. Data Mining: Concepts and Techniques (FH, S) (FH, ~S) (~FH, S) (~FH, ~S) LC 0.8 0.7 0.5 0.1 ~LC 0.2 0.5 0.3 0.9 Bayesian Belief Network: An Example Family History Smoker The conditional probability table (CPT) for variable LungCancer: LungCancer Emphysema CPT shows the conditional probability for each possible combination of its parents PositiveXRay Dyspnea Derivation of the probability of a particular combination of values of X, from CPT: Bayesian Belief Networks

47. Data Mining: Concepts and Techniques Training Bayesian Networks • Several scenarios: • Given both the network structure and all variables observable: learn only the CPTs • Network structure known, some hidden variables: gradient descent (greedy hill-climbing) method, analogous to neural network learning • Network structure unknown, all variables observable: search through the model space to reconstruct network topology • Unknown structure, all hidden variables: No good algorithms known for this purpose • Ref. D. Heckerman: Bayesian networks for data mining

48. Classification and Prediction • Types/ tools of classification and prediction • Classification by decision tree induction • Bayesian Classification • Classification by backpropagation • S.V.M. • Classification based on concepts from association rule mining • Other Classification Methods • Rough set • Fuzzy approach • Prediction • Classification accuracy • Summary

49. Soft Computing in Data Mining • Data Mining Techniques • Association techniques • Classification techniques • Classifiers • Bayesian classifier ( Naïve, Belief network) • Neural Network. • Genetic Algorithm • Rough set • Clustering • Clustering techniques • Data mining Process: • Data Mining Query Language (DMQL) • Soft Computing approach • Application domain • Biological data • Anomaly detection • Financial application • Issues and Challenges

50. Soft Computing approach • Soft computing is a consortium of methodologies that works synergistically and provides, in one form or another, flexibleinformation processing capability for handling real-life ambiguous situations . • Its aim is to exploit the tolerance for imprecision, uncertainty, approximate reasoning, and partial truth in order to achieve tractability, robustness, and low-cost solutions. • The guiding principle is to devise methods of computation that lead to an acceptable solution at low cost by seeking for an approximate solution to an imprecisely/precisely formulated problem.