CS 685 Special Topics in Data mining Instructor

CS 685 Special Topics in Data mining Instructor: Jinze Liu Spring 2008 The UNIVERSITY of KENTUCKY

Welcome! • Instructor: Jinze Liu Ø Homepage: http: //www. cs. uky. edu/~liuj Ø Office: 237 Hardymon Building Ø Email: liuj@cs. uky. edu Ø Office hour: by appointment 2 The UNIVERSITY of KENTUCKY

Overview • • Time: 2: 00 -3: 15 PM Tuesday and Thursday Place: POT 145 Credit: 3 Prerequisite: none Ø Preferred: Database, AI, Machine Learning, Statistics, Algorithms 3 The UNIVERSITY of KENTUCKY

Overview • Textbook: none Ø A collection of papers in recent conferences and journals • References Ø Data Mining --- Concepts and techniques, by Han and Kamber, Morgan Kaufmann, 2006. (ISBN: 1 -55860 -901 -6) Ø Introduction to Data Mining, by Tan, Steinbach, and Kumar, Addison Wesley, 2006. (ISBN: 0 -32136 -7) Ø Principles of Data Mining, by Hand, Mannila, and Smyth, MIT Press, 2001. (ISBN: 0 -262 -08290 -X) Ø The Elements of Statistical Learning --- Data Mining, Inference, and Prediction, by Hastie, Tibshirani, and Friedman, Springer, 2001. (ISBN: 0 -387 -95284 -5) Ø Mining the Web --- Discovering Knowledge from Hypertext Data, by Chakrabarti, Morgan Kaufmann, 2003. (ISBN: 155860 -754 -4) 4 The UNIVERSITY of KENTUCKY

Overview • Grading scheme Paper Presentation and discussion Project 40% 50% Attendance and 10% participation Ø No homework Ø No exam 5 The UNIVERSITY of KENTUCKY

Overview • Paper presentation Ø One per student Ø Research paper(s) v List of recommendations (will be available) v Your own pick (upon approval) Ø Three parts v Motivation for the research v Review of data mining methods v Discussion v Questions and comments from audience v Class participation: One question/comment per student Ø Order of presentation: will be arranged according to the topics Ø Please send in your choice of paper(s) by Jan 29 th. 6 The UNIVERSITY of KENTUCKY

Overview • Project (due May 1 st) Ø One project: Individual or team project Ø Some suggestion will be available shortly v You are welcome to propose your own especially you have a dataset for analysis. Ø Due Feb 7 th v Proposal: title and goal v Survey of related work: pros and cons v Outline of approach Ø Due April 1 st v Implementation update Ø Due May 1 st v Implementation v Evaluation v Discussion and future directions 7 The UNIVERSITY of KENTUCKY

Topics • Scope: Data Mining • Topics: Ø Ø Ø Ø Ø 8 Association Rule Sequential Patterns Graph Mining Clustering and Outlier Detection Classification and Prediction Regression Pattern Interestingness Dimensionality Reduction … The UNIVERSITY of KENTUCKY

Topics Ø Applications v. Biomedical informatics v. Bioinformatics v. Web mining v. Text mining v. Graphics v. Visualization v. Financial data analysis v. Intrusion detection v… 9 The UNIVERSITY of KENTUCKY

KDD References • Data mining and KDD (SIGKDD: CDROM) Ø Conferences: ACM-SIGKDD, IEEE-ICDM, SIAM-DM, PKDD, PAKDD, etc. Ø Journal: Data Mining and Knowledge Discovery, KDD Explorations • Database systems (SIGMOD: CD ROM) Ø Conferences: ACM-SIGMOD, ACM-PODS, VLDB, IEEE-ICDE, EDBT, ICDT, DASFAA Ø Journals: ACM-TODS, IEEE-TKDE, JIIS, J. ACM, etc. • AI & Machine Learning Ø Conferences: Machine learning (ICML), AAAI, IJCAI, COLT (Learning Theory), etc. Ø Journals: Machine Learning, Artificial Intelligence, etc. 10 The UNIVERSITY of KENTUCKY

KDD References • Statistics Ø Conferences: Joint Stat. Meeting, etc. Ø Journals: Annals of statistics, etc. • Bioinformatics Ø Conferences: ISMB, RECOMB, PSB, CSB, BIBE, etc. Ø Journals: J. of Computational Biology, Bioinformatics, etc. • Visualization Ø Conference proceedings: Info. Vis, CHI, ACM-SIGGraph, etc. Ø Journals: IEEE Trans. visualization and computer graphics, etc. 11 The UNIVERSITY of KENTUCKY