ALMA MATER STUDIORUM - UNIVERSITÀ DI BOLOGNA Data Mining M Laurea Magistrale in Ingegneria Informatica University of Bologna Course presentation Academic Year 2015/2016 Home page: http://www-db.disi.unibo.it/courses/DM/ Electronic version: 0.01.Presentation.pdf Electronic version: 0.01.Presentation-2p.pdf Bologna, September 23rd, 2015 Teachers Prof. Claudio Sartori e Prof.ssa Ilaria Bartolini Department of Computer Science and Engineering (DISI) Viale Risorgimento, 2 - 40136, Bologna Contacts E-mail: {claudio.sartori,i.bartolini}@unibo.it Telephone: 051 20 93554 (Sartori) - 051 20 93550 (Bartolini) Web: http://www-db.disi.unibo.it/csartori/ http://www-db.disi.unibo.it/ibartolini/ Office hours: Refer to personal Web page, c/o Palazziana DISI, close to entrance Via Vallescura (Prof. Sartori) on Fridays, from 16:00 to 18:00, c/o Palazziana DISI, close to entrance Via Vallescura (Prof.ssa Bartolini) I. Bartolini Data Mining M 2 General information Name: “Data Mining M” Credits: 8 Teaching hours: 64 hours Period: Semester I September 21st 2015 – December 18th 2015 Course organization: divided into two learning modules Module I – Data Mining (Prof. Sartori) From November 5th till December 17th 2015 Module II – Multimedia Data Mining (Prof.ssa Bartolini) Form September 23rd till November 4th 2015 I. Bartolini Data Mining M 3 Course calendar Teaching hours: Wednesday – 14:00-17:00 – Room 7.8 Scuole Sirani - Via Saragozza, 8 Thursday – 9:00-11:00 – Room 5.4 I. Bartolini Data Mining M 4 Course contents Learning outcomes The course aims to provide the students with the knowledge and skills necessary for the analysis of data in order to discover relationships and useful information for decisions support; particular attention is devoted to the presentation of the discovery process from the definition of the objectives and algorithms processing The second module provides a demonstration of how traditional data mining techniques can be profitably applied for the efficient management of multimedia collections in term of localization of data of interest and for purposes of visualization and browsing Parts of the course Process of knowledge discovery Data Mining techniques Multimedia data content representation Efficient and effective techniques for multimedia data retrieval, browsing, and visualization I. Bartolini Data Mining M 5 Program – Module I (Data Mining) Process of knowledge discovery Definition of objectives Selection of data sources Filtering, reconciliation and data transformation Data mining Validation and presentation of the results Data Mining techniques Classification with decision trees, neural networks and other algorithms Association rules Clustering/segmentation Analysis of case studies Examples with commercial data mining systems Architectures of systems with data mining components Standards for data mining components: PMML I. Bartolini Data Mining M 6 Program – Module II (Multimedia Data Mining) Multimedia data and content representations MM data and applications MM data coding MM data content representation How to find MM data of interest Description models for complex MM objects Similarity measures for MM data content MM Data Base Management Systems Efficient algorithms for MM information retrieval MM query formulation paradigms Sequential retrieval of MM data Index-based retrieval of MM data Automatic techniques for MM data semantic annotations Browsing MM data collections MM data presentation Result accuracy, use cases and real applications I. Bartolini Data Mining M 7 Course home page http://www-db.disi.unibo.it/courses/DM/ Contents: I. Bartolini News Copy of slides in PDF format Bibliography Software tools and useful links Assessment methods Exam sessions Topics for project activities Data Mining M 8 Readings/Bibliography Education material provided by the teachers (copies of the slides used in the classroom, scientific literature) Additional reading Tan, Steinbach, Kumar, "Introduction to Data Mining", Addison-Wesley, 2005. ISBN : 0321321367 Zhang, Zhang, "Multimedia Data Mining: A Systematic Introduction to Concepts and Theory", Chapman and Hall/CRC, 2008. ISBN: 9781584889663 I. Bartolini Data Mining M 9 Teaching methods Most course lectures are in "traditional" classrooms and exploit the slides Case studies are also proposed based on open-source software and frameworks I. Bartolini Data Mining M 10 Assessment methods The exam evaluation consists of an oral examination To participate to the exam, interested students have to register themselves by exploiting the usual UniBO Web application, called AlmaEsami The students can directly arrange with each teacher a Project Activity of Data Mining M based on their own preferences on provided topics (see Web page for more details) I. Bartolini Data Mining M 11 Examination sessions Six examination sessions per year divided as follows: three sessions during the winter starting from June, at the request of the students I. Bartolini Data Mining M 12