Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Course on Data Mining (581550-4) Intro/Ass. Rules 7.11. 24./26.10. Clustering 14.11. Episodes KDD Process Home Exam 30.10. Text Mining 28.11.2001 21.11. 28.11. Data mining - Applications, future, and summary Appl./Summary 1 Course on Data Mining (581550-4) Today 28.11.2001 • Today's subject: o Data mining applications, future, and summary • The program at the end of this week: o Exercise: KDD Process o Seminar: KDD Process 28.11.2001 Data mining - Applications, future, and summary 2 Applications, future and summary • Data mining applications • How to choose a data mining system? • Data mining system products and research prototypes • Additional themes on data mining • Social impact of data mining • Trends in data mining • Summary 28.11.2001 Data mining - Applications, future, and summary 3 Data mining applications • Data mining is a young discipline with wide and diverse applications o general principles of data mining versus domainspecific, effective data mining tools for particular applications • Application domains, e.g., o biomedical and DNA data analysis o financial data analysis o retail industry o telecommunication industry 28.11.2001 Data mining - Applications, future, and summary 4 Biomedical data mining and DNA analysis • DNA sequences consist of 4 basic building blocks (nucleotides): adenine (A), cytosine (C), guanine (G), and thymine (T). • Gene: a sequence of hundreds of individual nucleotides arranged in a particular order • Semantic integration of heterogeneous, distributed genome databases o data cleaning and data integration methods developed in data mining will help 28.11.2001 Data mining - Applications, future, and summary 5 DNA analysis – Examples (1) • Similarity search and comparison among DNA sequences o compare the frequently occurring patterns of each class o identify gene sequence patterns that play roles in various diseases • Association analysis: identification of co-occurring gene sequences o most diseases are triggered by a combination of genes acting together o may help determine the kinds of genes that are likely to co-occur together in target samples 28.11.2001 Data mining - Applications, future, and summary 6 DNA analysis – Examples (2) • Path analysis: linking genes to different disease development stages o different genes may become active at different stages of the disease o develop pharmaceutical interventions that target the different stages separately • Visualization tools and genetic data analysis 28.11.2001 Data mining - Applications, future, and summary 7 Data mining for financial data analysis (1) • Collected data is often relatively complete, reliable, and of high quality • Design and construction of data warehouses for multidimensional data analysis and data mining o view the debt and revenue changes, e.g., by month o access statistical information, e.g., trend • Loan payment prediction/consumer credit policy analysis o loan payment performance o consumer credit rating 28.11.2001 Data mining - Applications, future, and summary 8 Data mining for financial data analysis (2) • Classification and clustering of customers for targeted marketing o multidimensional segmentation to identify customer groups or associate a new customer to an appropriate customer group • Detection of money laundering and other financial crimes o integration of multiple DBs o tools: data visualization, linkage analysis, classification, clustering tools, outlier analysis, and sequential pattern analysis tools 28.11.2001 Data mining - Applications, future, and summary 9 Data mining for retail industry (1) • Retail industry: huge amounts of data on sales, customer shopping history, etc. • Applications of retail data mining: o identify customer buying behaviors o discover customer shopping patterns and trends o improve the quality of customer service o achieve better customer retention and satisfaction o enhance goods consumption ratios o design more effective goods transportation and distribution policies 28.11.2001 Data mining - Applications, future, and summary 10 Data mining in retail industry (2) • Design and construction of data warehouses based on the benefits of data mining (multidimensional analysis of sales, customers, products, time, and region) • Analysis of the effectiveness of sales campaigns • Analysis of customer loyalty o use customer loyalty card information to register sequences of purchases of particular customers o use sequential pattern mining to investigate changes in customer consumption or loyalty o suggest adjustments on the pricing and variety of goods • Purchase recommendation and cross-reference of items 28.11.2001 Data mining - Applications, future, and summary 11 Data mining for telecommunication industry (1) • A rapidly expanding and highly competitive industry and a great demand for data mining o understand the business involved o identify telecommunication patterns o catch fraudulent activities o make better use of resources o improve the quality of service • Multidimensional analysis of telecommunication data o e.g., calling-time, duration of call, location of caller, type of call, etc. 28.11.2001 Data mining - Applications, future, and summary 12 Data mining for telecommunication industry (2) • Fraudulent pattern analysis and the identification of unusual patterns o identify potentially fraudulent users and their atypical usage patterns o detect attempts to gain fraudulent entry to customer accounts o discover unusual patterns which may need special attention 28.11.2001 Data mining - Applications, future, and summary 13 Data mining for telecommunication industry (3) • Multidimensional association and sequential pattern analysis o find usage patterns for a set of communication services by customer group, by month, etc. o promote the sales of specific services o improve the availability of particular services in a region • Use of visualization tools in telecommunication data analysis 28.11.2001 Data mining - Applications, future, and summary 14 How to choose a data mining system? (1) • Commercial data mining systems have little in common o different data mining functionality or methodology o may even work with completely different kinds of data sets • For selection of a system we need to have a multiple dimensional view of existing systems 28.11.2001 Data mining - Applications, future, and summary 15 How to choose a data mining system? (2) • Data types: relational, transactional, text, time sequence, spatial? • System issues o running on only one or on several operating systems? o a client/server architecture? o provide Web-based interfaces and allow XML data as input and/or output? • Data sources o ASCII text files, multiple relational data sources o support ODBC connections (OLE DB, JDBC)? 28.11.2001 Data mining - Applications, future, and summary 16 How to choose a data mining system? (3) • Data mining functions and methodologies o one vs. multiple data mining functions o one vs. variety of methods per function • Coupling with DB and/or data warehouse systems o four forms of coupling: no coupling, loose coupling, semitight coupling, and tight coupling • Visualization tools: data visualization, mining result visualization, mining process visualization, and visual data mining 28.11.2001 Data mining - Applications, future, and summary 17 How to choose a data mining system? (4) • Scalability o row (or database size) scalability o column (or dimension) scalability o curse of dimensionality: it is much more challenging to make a system column scalable that row scalable • Data mining query language and graphical user interface o easy-to-use and high-quality graphical user interface o essential for user-guided, highly interactive data mining 28.11.2001 Data mining - Applications, future, and summary 18 Data mining systems (1) • IBM Intelligent Miner o a wide range of data mining algorithms o scalable mining algorithms o toolkits: neural network algorithms, statistical methods, data preparation, and data visualization tools o tight integration with IBM's DB2 relational database system • SAS Enterprise Miner o a variety of statistical analysis tools o data warehouse tools and multiple data mining algorithms 28.11.2001 Data mining - Applications, future, and summary 19 Data mining systems (2) • SGI MineSet o multiple data mining algorithms and advanced statistics o advanced visualization tools • Clementine (SPSS) o an integrated data mining development environment for end-users and developers o multiple data mining algorithms and visualization tools 28.11.2001 Data mining - Applications, future, and summary 20 Data mining systems (3) • DBMiner (DBMiner Technology Inc.) o multiple data mining modules: discovery-driven OLAP analysis, association, classification, and clustering o efficient, association and sequential-pattern mining functions, and visual classification tool o mining both relational databases and data warehouses • Microsoft SQLServer 2000 o integrate DB and OLAP with mining o support OLEDB for DM standard 28.11.2001 Data mining - Applications, future, and summary 21 Additional themes on data mining • • • • Web mining Visual data mining Audio data mining Theoretical foundations of data mining • Data mining and intelligent query answering 28.11.2001 Data mining - Applications, future, and summary 22 Web mining (1) • The WWW is huge, widely distributed, global information service center for o information services: news, advertisements, consumer information, education, government, e-commerce, etc. o hyper-link information o access and usage information 28.11.2001 Data mining - Applications, future, and summary 23 Web mining (2) • Web search engines: o index-based: search the Web, index Web pages, and build and store huge keyword-based indices o help locate sets of Web pages containing certain keywords • Deficiencies of the web search engines: o a topic of any breadth may easily contain hundreds of thousands of documents o many documents that are highly relevant to a topic may not contain keywords defining them 28.11.2001 Data mining - Applications, future, and summary 24 Web mining (3) • WWW provides rich sources for data mining • Challenges: o too huge for effective data warehousing and data mining o too complex and heterogeneous: no standards and structure 28.11.2001 Data mining - Applications, future, and summary 25 Web mining (4) • Web mining is a more challenging task than constructing and using web search engines • Web mining searches for o web access patterns o web structures o regularity and dynamics of web contents 28.11.2001 Data mining - Applications, future, and summary 26 Web mining (5) • Web mining taxonomy: Web Mining Web Content Mining Web Page Content Mining 28.11.2001 Web Structure Mining Search Result Mining Web Usage Mining General Access Pattern Tracking Data mining - Applications, future, and summary Customized Usage Tracking 27 Visual data mining (1) • Visualization: use of computer graphics to create visual images which aid in the understanding of complex, often massive representations of data • Visual data mining: the process of discovering implicit, but useful knowledge from large data sets using visualization techniques 28.11.2001 Data mining - Applications, future, and summary 28 Visual data mining (2) • Purpose of visualization o gain insight into an information space by mapping data onto graphical primitives o provide qualitative overview of large data sets o search for patterns, trends, structure, irregularities, relationships among data o help find interesting regions and suitable parameters for further quantitative analysis o provide a visual proof of computer representations derived 28.11.2001 Data mining - Applications, future, and summary 29 Visual data mining (3) • Integration of visualization and data mining o data visualization o data mining result visualization o data mining process visualization o interactive visual data mining 28.11.2001 Data mining - Applications, future, and summary 30 Data visualization • Data in a database or data warehouse can be viewed o at different levels of granularity or abstraction o as different combinations of attributes or dimensions • Data can be presented in various visual forms 28.11.2001 Data mining - Applications, future, and summary 31 Box-plots in Statsoft 28.11.2001 Data mining - Applications, future, and summary 32 Data mining result visualization • Presentation of the results or knowledge obtained from data mining in visual forms • Examples o scatter plots and box-plots o association rules o clusters o outliers o generalized rules 28.11.2001 Data mining - Applications, future, and summary 33 Scatter plots in SAS Enterprise Miner 28.11.2001 Data mining - Applications, future, and summary 34 Association rules in MineSet 3.0 28.11.2001 Data mining - Applications, future, and summary 35 A decision tree in MineSet 3.0 28.11.2001 Data mining - Applications, future, and summary 36 Cluster groupings in IBM Intelligent Miner 28.11.2001 Data mining - Applications, future, and summary 37 Data mining process visualization • Presentation of the various processes of data mining in visual forms so that users can see o how the data are extracted o from which database or data warehouse they are extracted o how the selected data are cleaned, integrated, preprocessed, and mined o which method is selected at data mining o where the results are stored o how they may be viewed 28.11.2001 Data mining - Applications, future, and summary 38 Data mining processes in Clementine 28.11.2001 Data mining - Applications, future, and summary 39 Interactive visual data mining • Using visualization tools in the data mining process to help users make smart data mining decisions • Example o display the data distribution in a set of attributes using colored sectors or columns o use the display to decide which sector should first be selected for classification and where a good split point for this sector may be 28.11.2001 Data mining - Applications, future, and summary 40 Interactive visual mining by perception-based classification 28.11.2001 Data mining - Applications, future, and summary 41 Audio data mining • Audio signals (sounds, music) are used to indicate the patterns of data, or the features of data mining results • An interesting alternative to visual mining • An inverse task of mining audio (such as music) databases which is to find patterns from audio data • Visual data mining may disclose interesting patterns using graphical displays, but requires users to concentrate on watching patterns • In audio data mining, the user listens to pitches, rhythms, tune, and melody in order to identify anything interesting or unusual 28.11.2001 Data mining - Applications, future, and summary 42 Theoretical foundations of data mining (1) • Data reduction o the basis of data mining is to reduce the data representation (use, e.g., histograms or clustering) o trades accuracy for speed • Data compression o the basis of data mining is compress the given data by encoding in terms of bits, association rules, decision trees, clusters, etc. 28.11.2001 Data mining - Applications, future, and summary 43 Theoretical foundations of data mining (2) • Pattern discovery o the basis of data mining is to discover patterns occurring in the database, e.g., associations, classification models and sequential patterns • Probability theory o the basis of data mining is to discover joint probability distributions of random variables 28.11.2001 Data mining - Applications, future, and summary 44 Theoretical foundations of data mining (3) • Microeconomic view o a view of utility o the task of data mining is finding patterns that are interesting only to the extent in that they can be used in the decision-making process of some enterprise 28.11.2001 Data mining - Applications, future, and summary 45 Theoretical foundations of data mining (4) • Inductive databases o data mining is the problem of performing inductive logic on databases o the task is to query the data and the theory (i.e., patterns) of the database o popular among many researchers in database systems 28.11.2001 Data mining - Applications, future, and summary 46 Data mining and intelligent query answering (1) • Query answering o direct query answering: returns exactly what is being asked o intelligent (or cooperative) query answering: analyzes the intent of the query and provides generalized, neighborhood or associated information relevant to the query 28.11.2001 Data mining - Applications, future, and summary 47 Data mining and intelligent query answering (2) • Some users may not have a clear idea of exactly what to mine or what is contained in the database • Intelligent query answering analyzes the user's intent and answers queries in an intelligent way 28.11.2001 Data mining - Applications, future, and summary 48 Data mining and intelligent query answering (3) • A general framework for the integration of data mining and intelligent query answering o data query: finds concrete data stored in a database o knowledge query: finds rules, patterns, and other kinds of knowledge in a database 28.11.2001 Data mining - Applications, future, and summary 49 Data mining and intelligent query answering (4) • For example, three ways to improve on-line shopping service o informative query answering by providing summary information o suggestion of additional items based on association analysis o product promotion by sequential pattern mining 28.11.2001 Data mining - Applications, future, and summary 50 Social impact of data mining • Is data mining a hype? • Data mining: merely managers’ business or everyone’s • Privacy and data security 28.11.2001 Data mining - Applications, future, and summary 51 Is data mining a hype, or will it be persistent? • Data mining is a technology • Technological life cycle: o innovators o early adopters o chasm o early majority o late majority o laggards 28.11.2001 Data mining - Applications, future, and summary 52 Life Cycle of Technology Adoption • Data mining is at chasm!? o existing data mining systems are too generic o need business-specific data mining solutions and smooth integration of business logic with data mining functions 28.11.2001 Data mining - Applications, future, and summary 53 Whose business is it? • Data mining will surely be an important tool for managers’ decision making • The amount of the available data is increasing, and data mining systems will be more affordable • Multiple personal uses o mine your family's medical history to identify genetically-related medical conditions o mine the records of the companies you deal with o mine data on stocks and company performance, etc. • Invisible data mining: build data mining functions into many intelligent tools 28.11.2001 Data mining - Applications, future, and summary 54 Threat to privacy and data security? • “Big Brother” is carefully watching you • Profiling information is collected constantly o you use your credit card, supermarket loyalty card, or frequent flyer card, or apply for any of the above o you surf the Web, reply to an Internet newsgroup, subscribe to a magazine, rent a video, or fill out a contest entry form • Collection of personal data may be beneficial for companies and consumers, but there is also potential for misuse 28.11.2001 Data mining - Applications, future, and summary 55 Protect privacy and data security • Fair information practices o international guidelines for data privacy protection o cover aspects relating to data collection, purpose, use, quality, openness, individual participation, and accountability o purpose specification and use limitation o openness: individuals have the right to know what information is collected about them, who has access to the data, and how the data are being used • Develop and use data security-enhancing techniques, e.g., blind signatures, biometric encryption, and anonymous databases 28.11.2001 Data mining - Applications, future, and summary 56 Trends in data mining (1) • Application exploration o development of application-specific data mining system o invisible data mining (mining as built-in function) • Scalable data mining methods o constraint-based mining: use of constraints to guide data mining systems in their search for interesting patterns 28.11.2001 Data mining - Applications, future, and summary 57 Trends in data mining (2) • Integration of data mining with database systems, data warehouse systems, and web database systems • Standardization of data mining language o a standard will facilitate systematic development, improve interoperability, and promote the education and use of data mining systems in industry and society • Visual data mining 28.11.2001 Data mining - Applications, future, and summary 58 Trends in data mining (3) • New methods for mining complex types of data o more research is required towards the integration of data mining methods with existing data analysis techniques for the complex types of data • Web mining • Privacy protection and information security in data mining 28.11.2001 Data mining - Applications, future, and summary 59 Summary (1) • Data mining: semi-automatic discovery of interesting patterns from large data sets • Knowledge discovery is a process: o preprocessing o data mining o postprocessing • Application areas: retail, telecommunication, Web mining, log analysis, … 28.11.2001 Data mining - Applications, future, and summary 60 Summary (2) • Knowledge can be mined from different kinds of databases (relational, object-oriented, spatial, WWW, …) • We can mine different kinds of knowledge (characterization, clustering, association, …) • Data mining uses also techniques from other areas of computer science (machine learning, statistics, visualization, …) 28.11.2001 Data mining - Applications, future, and summary 61 Summary (3) • Some useful data mining techniques: o association rules o episodes o text mining o classification o clustering • There are also many other data mining methods/techniques developed, but not covered in this course 28.11.2001 Data mining - Applications, future, and summary 62 Summary (4) • It is important to o study theoretical foundations of data mining o watch privacy and security issues in data mining • The future of data mining seems promising, even without hype 28.11.2001 Data mining - Applications, future, and summary 63 References - Applications etc. (1) • • • • • • • • • • • M. Ankerst, C. Elsen, M. Ester, and H.-P. Kriegel. Visual classification: An interactive approach to decision tree construction. KDD'99, San Diego, CA, Aug. 1999. P. Baldi and S. Brunak. Bioinformatics: The Machine Learning Approach. MIT Press, 1998. S. Benninga and B. Czaczkes. Financial Modeling. MIT Press, 1997. L. Breiman, J. Friedman, R. Olshen, and C. Stone. Classification and Regression Trees. Wadsworth International Group, 1984. M. Berthold and D. J. Hand. Intelligent Data Analysis: An Introduction. Springer-Verlag, 1999. M. J. A. Berry and G. Linoff. Mastering Data Mining: The Art and Science of Customer Relationship Management. John Wiley & Sons, 1999. A. Baxevanis and B. F. F. Ouellette. Bioinformatics: A Practical Guide to the Analysis of Genes and Proteins. John Wiley & Sons, 1998. Q. Chen, M. Hsu, and U. Dayal. A data-warehouse/OLAP framework for scalable telecommunication tandem traffic analysis. ICDE'00, San Diego, CA, Feb. 2000. W. Cleveland. Visualizing Data. Hobart Press, Summit NJ, 1993. S. Chakrabarti, S. Sarawagi, and B. Dom. Mining surprising patterns using temporal description length. VLDB'98, New York, NY, Aug. 1998. J. L. Devore. Probability and Statistics for Engineering and the Science, 4th ed. Duxbury Press, 1995. 28.11.2001 Data mining - Applications, future, and summary 64 References - Applications etc. (2) • • • • • • • • • • • A. J. Dobson. An Introduction to Generalized Linear Models. Chapman and Hall, 1990. B. Gates. Business @ the Speed of Thought. New York: Warner Books, 1999. M. Goebel and L. Gruenwald. A survey of data mining and knowledge discovery software tools. SIGKDD Explorations, 1:20-33, 1999. D. Gusfield. Algorithms on Strings, Trees and Sequences, Computer Science and Computation Biology. Cambridge University Press, New York, 1997. J. Han, Y. Huang, N. Cercone, and Y. Fu. Intelligent query answering by knowledge discovery techniques. IEEE Trans. Knowledge and Data Engineering, 8:373-390, 1996. R. C. Higgins. Analysis for Financial Management. Irwin/McGraw-Hill, 1997. C. H. Huberty. Applied Discriminant Analysis. New York: John Wiley & Sons, 1994. T. Imielinski and H. Mannila. A database perspective on knowledge discovery. Communications of ACM, 39:58-64, 1996. D. A. Keim and H.-P. Kriegel. VisDB: Database exploration using multidimensional visualization. Computer Graphics and Applications, pages 40-49, Sept. 94. J. M. Kleinberg, C. Papadimitriou, and P. Raghavan. A microeconomic view of data mining. Data Mining and Knowledge Discovery, 2:311-324, 1998. H. Mannila. Methods and problems in data mining. ICDT'99 Delphi, Greece, Jan. 1997. 28.11.2001 Data mining - Applications, future, and summary 65 References - Applications etc. (3) • • • • • • • • R. Mattison. Data Warehousing and Data Mining for Telecommunications. Artech House, 1997. R. G. Miller. Survival Analysis. New York: Wiley, 1981. G. A. Moore. Crossing the Chasm: Marketing and Selling High-Tech Products to Mainstream Customers. Harperbusiness, 1999. R. H. Shumway. Applied Statistical Time Series Analysis. Prentice Hall, 1988. E. R. Tufte. The Visual Display of Quantitative Information. Graphics Press, Cheshire, CT, 1983. E. R. Tufte. Envisioning Information. Graphics Press, Cheshire, CT, 1990. E. R. Tufte. Visual Explanations : Images and Quantities, Evidence and Narrative. Graphics Press, Cheshire, CT, 1997. M. S. Waterman. Introduction to Computational Biology: Maps, Sequences, and Genomes (Interdisciplinary Statistics). CRC Press, 1995. 28.11.2001 Data mining - Applications, future, and summary 66 Data mining conferences • • • • • • 1989 IJCAI Workshop 1991-1994 KDD Workshops 1995-1998 KDD Conferences 1998 ACM SIGKDD 1999-> SIGKDD Conferences And many smaller/new DM conferences, e.g., o PAKDD, PKDD o SIAM-Data Mining, (IEEE) ICDM 28.11.2001 Data mining - Applications, future, and summary 67 Useful References on Data Mining • DM: o Conferences: KDD, PKDD, PAKDD, ... o Journals: Data Mining and Knowledge Discovery, CACM • DM/DB: o Conferences: ACM-SIGMOD/PODS, VLDB, ... o Journals: ACM-TODS, J. ACM, IEEE-TKDE, JIIS, ... • AI/ML: o Conferences: Machine Learning, AAAI, IJCAI, ... o Journals: Machine Learning, Artifical Intelligence, ... 28.11.2001 Data mining - Applications, future, and summary 68 Reminder: Course Organization Course Evaluation • • 28.11.2001 Passing the course: min 30 points o home exam: min 13 points (max 30 points) o exercises/experiments: min 8 points (max 20 points) at least 3 returned and reported experiments o group presentation: min 4 points (max 10 points) Remember also the other requirements: o attending the lectures (5/7) o attending the seminars (4/5) o attending the exercises (4/5) Data mining - Applications, future, and summary 69 Data mining applications, future, and summary Thanks to Jiawei Han from Simon Fraser University for his slides which greatly helped in preparing this lecture! 28.11.2001 Data mining - Applications, future, and summary 70