- Количество слайдов: 14
Comp 3503 Knowledge Discovery and Data Mining Daniel L. Silver, Ph. D.
Comp 5013 Machine Learning and Data Mining Daniel L. Silver, Ph. D.
Outline Who am I? n Objectives of the course n Review of the course homepage n Stuff you need to have and do n 19 -Mar-18 Daniel L. Silver 3
Who am I? n n Danny Silver - BSc(Acadia), MSc, Ph. D (UWO) Background/Experience: – 14 years industry experience: » 2 years N. S. Government (Systems Programmer) » 9 years MTT – MIS (Prog. - Project Manager/Advisor) » 3 years SHL System House (Tech Architect, Project Manager) – 3 years Dalhousie (1996 -1999) – Started at Acadia in 1999 – 20 years Cog. Nova Technologies (Private Consulting) n The Bad News and the Good News 19 -Mar-18 Daniel L. Silver 4
Who are you? Name 19 -Mar-18 Course Interest Daniel L. Silver 5
Objectives of 3503 n To introduce the processes, theory and technologies of Data Analytics: – – n collection, cleaning and consolidation of data conversion of data into information dissemination of that information for the generation of human knowledge. Key discussion areas: – – – Data/Knowledge Management Knowledge Discovery Process Data Warehousing Data Mining Data Visualization 19 -Mar-18 Daniel L. Silver 6
Objectives of 3503 n By the end of the course you will understand: – Knowledge discovery (data analytics) process and its major activities, and management issues – Differences and relationships between deductive hypothesis-driven discovery and inductive data-driven modeling – Fundamentals of data warehousing, data mining and data visualization – Fundamentals of supervised and unsupervised learning – Major management and technical issues surrounding data security and privacy – Have hands-on experience with statistical, data mining, and data visualization software 19 -Mar-18 Daniel L. Silver 7
Objectives of 5013 n n To introduce the processes, theory and technologies of Data Analytics (KDD and DM) To provide fundamental theory of machine learning To provide experience at developing and testing ML software Key learning areas: – – – Supervised learning Unsupervised learning Semi-supervised methods Deep learning architectures Reinforcement learning (if time allows) 19 -Mar-18 Daniel L. Silver 8
Joint Structure of Courses n There is no TA n 1: 30 -3: 00 pm on Tues/Thur: – 3503 classes – Joint classes for common material n 4: 30 -6: 00 pm on Tues/Thurs: – 5013 classes – Joint tutorials 19 -Mar-18 Daniel L. Silver 9
Review 3503 course homepage
Review 5013 course homepage
Stuff you will need to have Text books (see websites) n Tech Services compliant laptop n Software: n – MS Office or Open Office suite – Weka Data Mining environment (Mac, Win) – Ward Systems Group NS 2 (Windows only) – 3503: IBM Cognos Insight (Windows only) – 5013: C, Java, Matlab programming environ. 19 -Mar-18 Daniel L. Silver 12
Stuff you will need to do n Come to class – Deeper discussion of issues – Handouts – Quizzes n Come to class prepared – Read material in advance – Be prepared to answer and ask questions 19 -Mar-18 Daniel L. Silver 13
THE END danny. [email protected] ca