1. 1CISC 4631: Data Mining Fall 2010 Introduction to Data Mining (these slides are based on a variety of sources)

2. Let’s Start By Seeing What you Know Quick Quiz Quick Quiz Do you know what Data Mining is? Do you know what Data Mining is? Do you know of any examples of Data Mining? Do you know of any examples of Data Mining? 2CISC 4631: Data Mining

3. 3 What is Data Mining? Data Mining has many definitions Data Mining has many definitions Non-trivial extraction of implicit, previously unknown and potentially useful information from data Non-trivial extraction of implicit, previously unknown and potentially useful information from data Exploration & analysis, by automatic or semi-automatic means, of large quantities of data in order to discover meaningful patterns Exploration & analysis, by automatic or semi-automatic means, of large quantities of data in order to discover meaningful patterns

4. 4CSRU4631: Data Mining Alternative Names Data Mining also known as or related to: Data Mining also known as or related to: Knowledge discovery in databases (KDD) Knowledge discovery in databases (KDD) Knowledge extraction Knowledge extraction Data/pattern analysis Data/pattern analysis Data archeology, data dredging, information harvesting, business intelligence, etc. Data archeology, data dredging, information harvesting, business intelligence, etc.

5. Some Examples Netflix and Amazon use data mining to recommend products (recommender systems) Netflix and Amazon use data mining to recommend products (recommender systems) Companies use data mining for marketing Companies use data mining for marketing Who should be mailed a catalog Who should be mailed a catalog Who should see what online ads (Google Adwords) Who should see what online ads (Google Adwords) My WISDM project uses data mining to determine (from your cell phone accelerometer data) who you are and what you are doing My WISDM project uses data mining to determine (from your cell phone accelerometer data) who you are and what you are doing 5CISC 4631: Data Mining

6. Why Data Mining and Why Now? Data Mining was not very popular until the last 10 years or so. Data Mining was not very popular until the last 10 years or so. Quick Quiz: Quick Quiz: Why is it data mining popular now? Why is it data mining popular now? What changed? What changed? 6CISC 4631: Data Mining

7. Why Mine Data? There are now tremendous amounts of data that are automatically collected and warehoused. What are some examples? There are now tremendous amounts of data that are automatically collected and warehoused. What are some examples? Web data, e-commerce Web data, e-commerce Store purchases Store purchases Bank/Credit Card transactions Bank/Credit Card transactions Cell phone GPS information Cell phone GPS information 7CISC 4631: Data Mining

8. Why Mine Data? What technological changes have helped make data mining so prevalent now? What technological changes have helped make data mining so prevalent now? Computers: cheaper and more powerful Computers: cheaper and more powerful Smaller mobile devices are exploding in popularity Smaller mobile devices are exploding in popularity Disk and other storage: greater capacity and cheaper Disk and other storage: greater capacity and cheaper RFID (radio frequency IDs), bar codes, etc RFID (radio frequency IDs), bar codes, etc Increased use of on-line resources and Internet Increased use of on-line resources and Internet 8CISC 4631: Data Mining

9. Why Mine Data? In business, competitive pressure is strong In business, competitive pressure is strong Provide better, customized services for an edge (e.g. in Customer Relationship Management) Provide better, customized services for an edge (e.g. in Customer Relationship Management) CRM is a relatively big deal now CRM is a relatively big deal now How do we get the most out of the customer over the long run How do we get the most out of the customer over the long run Example: Customer Churn Analysis Example: Customer Churn Analysis 9CISC 4631: Data Mining

10. Scientific Viewpoint Data collected at enormous speeds Data collected at enormous speeds remote sensors on satellite remote sensors on satellite telescopes scanning the skies telescopes scanning the skies microarrays generating gene expression data microarrays generating gene expression data scientific simulations scientific simulations Traditional techniques infeasible Traditional techniques infeasible Data mining may help scientists Data mining may help scientists in classifying and segmenting data in classifying and segmenting data in hypothesis formation in hypothesis formation 10CISC 4631: Data Mining

11. 11CISC 4631: Data Mining Mining Large Data Sets - Motivation Often information “ hidden ” in data that is not evident Often information “ hidden ” in data that is not evident Human analysts may take weeks to discover useful info Human analysts may take weeks to discover useful info Much of the data is never analyzed at all Much of the data is never analyzed at all The Data Gap Total new disk (TB) since 1995 # of analysts From: R. Grossman, C. Kamath, V. Kumar, “Data Mining for Scientific and Engineering Applications”

12. 12CISC 4631: Data Mining Mining Large Data Sets - Motivation AT&T’s 26TB call detail database (2003) AT&T’s 26TB call detail database (2003) Ebay 6PB, IRS 150TB data warehouse Ebay 6PB, IRS 150TB data warehouse Yahoo has a 2PB DB to analyze behavior of ½ billion web visitors/month (24 billion events/day) Yahoo has a 2PB DB to analyze behavior of ½ billion web visitors/month (24 billion events/day) Wal-Mart has a 583 TB database (2006) Wal-Mart has a 583 TB database (2006) Indexed web contains about 20 Billion pages Indexed web contains about 20 Billion pages Sites like Facebook, Flicker & Twitter contain lots of data Sites like Facebook, Flicker & Twitter contain lots of data

13. 13CISC 4631: Data Mining electronic point-of-sale data hospital patient registries catalog orders bank transactions remote sensing images tax returns airline reservations credit card charges stock trades OLTP telephone calls Data Deluge electronic point-of-sale data hospital patient registries stock trades OLTP telephone calls electronic point-of-sale data hospital patient registries catalog orders bank transactions stock trades OLTP telephone calls electronic point-of-sale data hospital patient registries remote sensing images tax returns catalog orders bank transactions stock trades OLTP telephone calls electronic point-of-sale data hospital patient registries airline reservations credit card charges remote sensing images tax returns catalog orders bank transactions stock trades OLTP telephone calls electronic point-of-sale data hospital patient registries

14. 14CISC 4631: Data Mining Amount of Data Created in One Year Humans created/copied 161/281 Exabytes in 06/07 (IDC) Humans created/copied 161/281 Exabytes in 06/07 (IDC) 1 Exabyte = 10 18 1 Exabyte = 10 18 12 stacks of books stretching from Earth to Sun 12 stacks of books stretching from Earth to Sun 3 million times the books ever written 3 million times the books ever written In 2010 will be 988 Exabytes In 2010 will be 988 Exabytes Not all data stored at once Not all data stored at once Much only temporarily Much only temporarily UC Berkeley 2003 estimate: UC Berkeley 2003 estimate: 5 Exabytes of new data created in 2002 5 Exabytes of new data created in 2002 US produces ~40% of new stored data worldwide US produces ~40% of new stored data worldwide

15. 15CISC 4631: Data Mining Data Growth Rate Twice as much information was created in 2002 as in 1999 (~30% growth rate) Twice as much information was created in 2002 as in 1999 (~30% growth rate) Other growth rate estimates even higher Other growth rate estimates even higher Very little data will ever be looked at by a human Very little data will ever be looked at by a human Knowledge Discovery is NEEDED to make sense and use of data Knowledge Discovery is NEEDED to make sense and use of data Moore’s Law: Moore’s Law: The information density on silicon-integrated circuits doubles every 18 to 24 months. The information density on silicon-integrated circuits doubles every 18 to 24 months. Parkinson’s Law: Parkinson’s Law: Work expands to fill the time available for its completion Work expands to fill the time available for its completion

16. 16CISC 4631: Data Mining Draws ideas from machine learning/AI, pattern recognition, statistics, and database systems Draws ideas from machine learning/AI, pattern recognition, statistics, and database systems Traditional techniques may be unsuitable due to Traditional techniques may be unsuitable due to Enormity of data Enormity of data High dimensionality of data High dimensionality of data Heterogeneous, distributed nature of data Heterogeneous, distributed nature of data Origins of Data Mining Artificial Intelligence / Machine Learning/ Pattern Recognition Statistics Data Mining Database systems

17. 17CISC 4631: Data Mining Origins of Data Mining: My view Biggest contributor is Machine Learning, which is a subfield of Artificial Intelligence Biggest contributor is Machine Learning, which is a subfield of Artificial Intelligence Data Mining is a subset of machine learning and focuses on practical problems of learning from data Data Mining is a subset of machine learning and focuses on practical problems of learning from data Unlike machine learning, ultimate goal is not to build something that can learn as flexibly as a human Unlike machine learning, ultimate goal is not to build something that can learn as flexibly as a human Does include other data analysis methods, like statistics Does include other data analysis methods, like statistics Databases do not play a central role in data mining. Databases do not play a central role in data mining. Most DM does not occur on data in a conventional database, but rather extracts it to a flat file. Most DM does not occur on data in a conventional database, but rather extracts it to a flat file. Data Mining methods do not work while data in a conventional (relational) database. Data Mining methods do not work while data in a conventional (relational) database.

18. 18CISC 4631: Data Mining Statistics & Machine Learning vs. Data Mining When compared to Data Mining: When compared to Data Mining: Statistics is: Statistics is: more theory-based/based on mathematics as opposed to heuristic methods more theory-based/based on mathematics as opposed to heuristic methods more focused on testing hypotheses more focused on testing hypotheses makes more assumptions about the data makes more assumptions about the data Machine learning is: Machine learning is: focused on improving performance of a learning agent in an environment focused on improving performance of a learning agent in an environment

19. 19CISC 4631: Data Mining The KDD (Data Mining) Process Data Mining is a process, sometimes referred to as a knowledge discovery process. In this process there is a data mining step that applies data mining algorithms to extract knowledge. About 80% of our class in on the data mining step.

20. 20CISC 4631: Data Mining Back to “What is a Data Mining”? My opinion: My opinion: Before determining whether something is data mining need to consider: Before determining whether something is data mining need to consider: Is it a DM task? Is it a DM task? Is it implemented using a DM method? Is it implemented using a DM method? Ideally, both parts will use data mining but may be considered DM even if only is used for one. Ideally, both parts will use data mining but may be considered DM even if only is used for one. We now will list the key DM tasks We now will list the key DM tasks The course is organized around these tasks The course is organized around these tasks

21. DATA MINING TASKS Second Part of Introduction: 21CISC 4631: Data Mining

22. 22CISC 4631: Data Mining 2 Top Level Data Mining Tasks At highest level, data mining tasks can be divided into: At highest level, data mining tasks can be divided into: Prediction Tasks Prediction Tasks Use some variables to predict unknown or future values of other variables Use some variables to predict unknown or future values of other variables Description Tasks Description Tasks Find human-interpretable patterns that describe the data Find human-interpretable patterns that describe the data

23. 23CISC 4631: Data Mining Key Data Mining Tasks Overview of major data mining tasks: Overview of major data mining tasks: Predictive Predictive Classification Classification Regression Regression Deviation/Anomaly Detection Deviation/Anomaly Detection Descriptive Descriptive Clustering Clustering Association Rule Discovery Association Rule Discovery Sequential Pattern Discovery Sequential Pattern Discovery

24. 24CISC 4631: Data Mining Classification: Definition Given a collection of records (training set ) Given a collection of records (training set ) Each record contains a set of attributes, one of the attributes is the class, which is to be predicted. Each record contains a set of attributes, one of the attributes is the class, which is to be predicted. Find a model for class attribute as a function of the values of other attributes. Find a model for class attribute as a function of the values of other attributes. Model maps record to a class value Model maps record to a class value Goal: previously unseen records should be assigned a class as accurately as possible. Goal: previously unseen records should be assigned a class as accurately as possible. A test set is used to determine the accuracy of the model. A test set is used to determine the accuracy of the model. Can you think of examples of classification tasks? We will see several shortly. Can you think of examples of classification tasks? We will see several shortly.

25. 25CISC 4631: Data Mining Classification Example categorical continuous class Test Set Training Set Model Learn Classifier Task: Predict if someone cheats on their taxes

26. 26CISC 4631: Data Mining Classification: Application 1 Direct Marketing Direct Marketing Goal: Reduce cost of mailing by targeting a set of consumers likely to buy a new cell-phone product. Goal: Reduce cost of mailing by targeting a set of consumers likely to buy a new cell-phone product. Approach: Approach: Use the data for a similar product introduced before. Use the data for a similar product introduced before. We know which customers decided to buy and which decided otherwise. This {buy, don’t buy} decision forms the class attribute We know which customers decided to buy and which decided otherwise. This {buy, don’t buy} decision forms the class attribute Collect various demographic, lifestyle, and company- interaction related information about all such customers. Collect various demographic, lifestyle, and company- interaction related information about all such customers. Type of business, where they stay, how much they earn, etc. Type of business, where they stay, how much they earn, etc. Use this info as input attributes to learn a classifier model Use this info as input attributes to learn a classifier model

27. 27CISC 4631: Data Mining Classification: Application 2 Fraud Detection Fraud Detection Goal: Predict fraudulent cases in credit card transactions Goal: Predict fraudulent cases in credit card transactions Approach: Approach: Use credit card transactions and info on account-holders as attributes Use credit card transactions and info on account-holders as attributes When and what does customer buy, how often pays on time, etc When and what does customer buy, how often pays on time, etc Label past transactions as fraud or fair transactions. This forms the class attribute. Label past transactions as fraud or fair transactions. This forms the class attribute. Learn a model for the class of the transactions. Learn a model for the class of the transactions. Use this model to detect fraud by observing credit card transactions on an account. Use this model to detect fraud by observing credit card transactions on an account.

28. 28CISC 4631: Data Mining Classification: Application 3 Customer Attrition/Churn: Customer Attrition/Churn: Goal: To predict whether a customer is likely to be lost to a competitor. Goal: To predict whether a customer is likely to be lost to a competitor. Approach: Approach: Use detailed record of transactions with each of the past and present customers, to find attributes. Use detailed record of transactions with each of the past and present customers, to find attributes. How often the customer calls, where he calls, what time-of-the day he calls most, his financial status, marital status, etc. How often the customer calls, where he calls, what time-of-the day he calls most, his financial status, marital status, etc. Label the customers as loyal or disloyal. Label the customers as loyal or disloyal. Find a model for loyalty. Find a model for loyalty.

29. 29CISC 4631: Data Mining Classification: Application 4 Sky Survey Cataloging Sky Survey Cataloging Goal: To predict class (star or galaxy) of sky objects, especially visually faint ones, based on the telescopic survey images (from Palomar Observatory). Goal: To predict class (star or galaxy) of sky objects, especially visually faint ones, based on the telescopic survey images (from Palomar Observatory). 3000 images with 23,040 x 23,040 pixels per image. 3000 images with 23,040 x 23,040 pixels per image. Approach: Approach: Segment the image. Segment the image. Measure image attributes (features) - 40 of them per object. Measure image attributes (features) - 40 of them per object. Model the class based on these features. Model the class based on these features. Success Story: Could find 16 new high red-shift quasars, some of the farthest objects that are difficult to find! Success Story: Could find 16 new high red-shift quasars, some of the farthest objects that are difficult to find! From [Fayyad, et.al.] Advances in Knowledge Discovery and Data Mining, 1996

30. 30CISC 4631: Data Mining Classifying Galaxies Early Intermediate Late Data Size: 72 million stars, 20 million galaxies Object Catalog: 9 GB Image Database: 150 GB Class: Stages of Formation Attributes: Image features, Characteristics of light waves received, etc. Courtesy: http://aps.umn.edu

31. 31CISC 4631: Data Mining Regression Predict a value of a given continuous valued variable based on the values of other variables, assuming a linear or nonlinear model of dependency. Predict a value of a given continuous valued variable based on the values of other variables, assuming a linear or nonlinear model of dependency. Greatly studied in statistics, neural network fields. Greatly studied in statistics, neural network fields. Examples: Examples: Predicting sales amounts of new product based on advertising expenditure. Predicting sales amounts of new product based on advertising expenditure. Predicting wind velocities as a function of temperature, humidity, air pressure, etc. Predicting wind velocities as a function of temperature, humidity, air pressure, etc. Time series prediction of stock market indices. Time series prediction of stock market indices.

32. 32CISC 4631: Data Mining Clustering Given a set of data points, each having a set of attributes, and a similarity measure among them, find clusters such that Given a set of data points, each having a set of attributes, and a similarity measure among them, find clusters such that Data points in one cluster are similar to one another Data points in one cluster are similar to one another Data points in different clusters are not (less) similar Data points in different clusters are not (less) similar Similarity Measures: Similarity Measures: Euclidean distance if attributes are continuous Euclidean distance if attributes are continuous Problem-specific measures Problem-specific measures Can you think of any applications of clustering? Can you think of any applications of clustering?

33. 33CISC 4631: Data Mining Illustrating Clustering x Euclidean Distance Based Clustering in 3-D space. Intracluster distances are minimized Intracluster distances are minimized Intercluster distances are maximized Intercluster distances are maximized

34. 34CISC 4631: Data Mining Clustering: Application 1 Market Segmentation: Market Segmentation: Goal: subdivide a market into distinct subsets of customers where any subset may conceivably be selected as a market target to be reached with a distinct marketing mix. Goal: subdivide a market into distinct subsets of customers where any subset may conceivably be selected as a market target to be reached with a distinct marketing mix. Approach: Approach: Collect different attributes of customers based on their geographical and lifestyle related information. Collect different attributes of customers based on their geographical and lifestyle related information. Find clusters of similar customers. Find clusters of similar customers. Measure the clustering quality by observing buying patterns of customers in same cluster vs. those from different clusters. Measure the clustering quality by observing buying patterns of customers in same cluster vs. those from different clusters.

35. 35CISC 4631: Data Mining Clustering: Application 2 Document Clustering: Document Clustering: Goal: To find groups of documents that are similar to each other based on the important terms appearing in them. Goal: To find groups of documents that are similar to each other based on the important terms appearing in them. Approach: To identify frequently occurring terms in each document. Form a similarity measure based on the frequencies of different terms. Use it to cluster. Approach: To identify frequently occurring terms in each document. Form a similarity measure based on the frequencies of different terms. Use it to cluster. Gain: Information Retrieval can utilize the clusters to relate a new document or search term to clustered documents. Gain: Information Retrieval can utilize the clusters to relate a new document or search term to clustered documents.

36. 36CISC 4631: Data Mining Association Rule Discovery Given a set of records each of which contain some number of items from a given collection Given a set of records each of which contain some number of items from a given collection Produce dependency rules which will predict occurrence of an item based on occurrences of other items. Produce dependency rules which will predict occurrence of an item based on occurrences of other items. Rules Discovered: {Milk} --> {Coke} {Diaper, Milk} --> {Beer} Rules Discovered: {Milk} --> {Coke} {Diaper, Milk} --> {Beer}

37. 37CISC 4631: Data Mining Association Rule Discovery: Application 1 Marketing and Sales Promotion: Marketing and Sales Promotion: Let the rule discovered be Let the rule discovered be {Bagels, … } --> {Potato Chips} {Bagels, … } --> {Potato Chips} Potato Chips as consequent => Can be used to determine what should be done to boost its sales. Potato Chips as consequent => Can be used to determine what should be done to boost its sales. Bagels in the antecedent => C an be used to see which products would be affected if the store discontinues selling bagels. Bagels in the antecedent => C an be used to see which products would be affected if the store discontinues selling bagels. Bagels in antecedent and Potato chips in consequent => Can be used to see what products should be sold with Bagels to promote sale of Potato chips! Bagels in antecedent and Potato chips in consequent => Can be used to see what products should be sold with Bagels to promote sale of Potato chips! Can help determine where to position store items Can help determine where to position store items

38. 38CISC 4631: Data Mining Association Rule Discovery: Application 2 Supermarket shelf management Supermarket shelf management Goal: Identify items that are bought together by many customers Goal: Identify items that are bought together by many customers Approach: Process the point-of-sale data collected with barcode scanners to find item dependencies Approach: Process the point-of-sale data collected with barcode scanners to find item dependencies A “classic” rule -- A “classic” rule -- If a customer buys diaper and milk, then he is very likely to buy beer. If a customer buys diaper and milk, then he is very likely to buy beer.

39. 39CISC 4631: Data Mining Sequential Pattern Discovery: Definition Given is a set of objects, with each object associated with its own timeline of events, find rules that predict strong sequential dependencies among different events. Given is a set of objects, with each object associated with its own timeline of events, find rules that predict strong sequential dependencies among different events. Rules are formed by first disovering patterns. Event occurrences in the patterns are governed by timing constraints. Rules are formed by first disovering patterns. Event occurrences in the patterns are governed by timing constraints. (A B) (C) (D E)

40. 40CISC 4631: Data Mining Sequential Pattern Discovery: Examples In telecommunications alarm logs, In telecommunications alarm logs, (Inverter_Problem Excessive_Line_Current) (Inverter_Problem Excessive_Line_Current) (Rectifier_Alarm) --> (Fire_Alarm) (Rectifier_Alarm) --> (Fire_Alarm) In point-of-sale transaction sequences, In point-of-sale transaction sequences, Computer Bookstore: Computer Bookstore: (Intro_To_Visual_C) (C++_Primer) --> (Perl_for_dummies,Tcl_Tk) (Intro_To_Visual_C) (C++_Primer) --> (Perl_for_dummies,Tcl_Tk) Athletic Apparel Store: Athletic Apparel Store: (Shoes) (Racket, Racketball) --> (Sports_Jacket) (Shoes) (Racket, Racketball) --> (Sports_Jacket)

41. 41CISC 4631: Data Mining Deviation/Anomaly Detection Detect significant deviations from normal behavior Detect significant deviations from normal behavior Applications: Applications: Credit Card Fraud Detection Credit Card Fraud Detection Network Intrusion Detection Network Intrusion Detection Typical network traffic at University level may reach over 100 million connections per day

42. 42CISC 4631: Data Mining Challenges of Data Mining Scalability Scalability Dimensionality Dimensionality Complex and Heterogeneous Data Complex and Heterogeneous Data Data Quality Data Quality Data Ownership and Distribution Data Ownership and Distribution Privacy Preservation Privacy Preservation Streaming Data Streaming Data

43. 43CISC 4631: Data Mining What is (and is not) Data Mining? Based on the definitions of data mining, are these DM or not? Based on the definitions of data mining, are these DM or not? Finding a phone number in a directory Finding a phone number in a directory Not data mining (trivial) Not data mining (trivial) Grouping related documents returned by search engine Grouping related documents returned by search engine Is data mining Is data mining Identifying who has a disease based on symptoms Identifying who has a disease based on symptoms Is data mining (not trivial) Is data mining (not trivial) Web search on keyword using search engine Web search on keyword using search engine May be data mining** May be data mining** ** More of an information retrieval task than data mining task, but since a search engine like Google does more than just keyword matching– it decides which web pages are important or not (a classification task that is part of DM) in order to get good results, the answer is not clear.

44. 44CISC 4631: Data Mining If you are Interested in Data Mining Visit kdnuggets, an online newsletter and more Visit kdnuggets, an online newsletter and more http://www.kdnuggets.com http://www.kdnuggets.com http://www.kdnuggets.com You can arrange to have newsletter emailed to you You can arrange to have newsletter emailed to you Also includes job openings Also includes job openings ACM SIGKDD is the professional organization associated with data mining ACM SIGKDD is the professional organization associated with data mining ACM Special Interest Group (SIG) on data mining ACM Special Interest Group (SIG) on data mining Can join SIGKDD for $22 or for $54 can also join ACM as student member Can join SIGKDD for $22 or for $54 can also join ACM as student member Conferences Conferences KDD, ICDM, DMIN, … KDD, ICDM, DMIN, …

45. 45CISC 4631: Data Mining Course Projects Projects must involve data mining Projects must involve data mining May be research related May be research related Examine some aspect of data mining Examine some aspect of data mining May be application oriented May be application oriented Solve a realistic, complex, problem Solve a realistic, complex, problem May be a combination of both May be a combination of both Most problems involve some interesting aspect Most problems involve some interesting aspect In some cases can be a survey/analysis paper (i.e., just a report), but this will be atypical In some cases can be a survey/analysis paper (i.e., just a report), but this will be atypical Can be done individually or in teams of 2 Can be done individually or in teams of 2 Ideally some projects can be published in a workshop or conference Ideally some projects can be published in a workshop or conference

46. 46CISC 4631: Data Mining Course Projects Output Output A written report, similar to a workshop or conference paper A written report, similar to a workshop or conference paper Two example workshop papers from last time course offered : Two example workshop papers from last time course offered : http://storm.cis.fordham.edu/~gweiss/papers/ubdm05-mccarthy.pdf http://storm.cis.fordham.edu/~gweiss/papers/ubdm05-mccarthy.pdf http://storm.cis.fordham.edu/~gweiss/papers/ubdm05-mccarthy.pdf http://storm.cis.fordham.edu/~gweiss/papers/ubdm05-ciraco.pdf http://storm.cis.fordham.edu/~gweiss/papers/ubdm05-ciraco.pdf http://storm.cis.fordham.edu/~gweiss/papers/ubdm05-ciraco.pdf For more examples: For more examples: http://storm.cis.fordham.edu/~gweiss/publications.html and look at the various workshop/conference papers http://storm.cis.fordham.edu/~gweiss/publications.html and look at the various workshop/conference papers http://storm.cis.fordham.edu/~gweiss/publications.html A presentation in class near end of semester A presentation in class near end of semester Stretch goal: submit paper to a workshop or conference Stretch goal: submit paper to a workshop or conference I can help you I can help you

47. 47CISC 4631: Data Mining Course Projects The sooner you start the better The sooner you start the better Think about: Think about: What you know about What you know about What data you have access to What data you have access to What type of problems you are interested in What type of problems you are interested in Who you want to work with Who you want to work with I will provide some specific project ideas I will provide some specific project ideas Areas include: Areas include: Classification, clustering, association rules Classification, clustering, association rules Web and link mining, text mining, social network analysis Web and link mining, text mining, social network analysis