MSc Data Mining

BrightSpace : This module is now available on BrightSpace. Each student needs to self-enroll in the module. Search for ‘Data Mining DATA9900‘, and for my name. The module description will have my name. This Brightspace module will be used for TU59 and TU60.  Look for sub-sections for links for online classes.  Online classes for TU59 will be listed in module subsection called TU59 Online Classes. Similarly for TU60.

Students are reminded that notes provided on this site are intended to form summary material only and are not intended to be a substitute for attending lectures or further reading on the subject.

My slides are not Lecture Notes

Students should download the notes to your own device. The notes are a living artifact and will evolve from semester to semester. It cannot be guaranteed that the notes will be available after the end of a semester.

Important: You will need to perform additional reading and research (outside of class time) to complete the various lab exercises and to explore the topics covered in more detail.

Important: The notes are living artifacts. I may make some minor changes to these. The videos are pre-recorded and may not reflect the current version of the notes.

*** Module Introduction & Admin – Notes

FAQ for module – Please check regularly – See new Updates about Assignment

Week of/begin
 Lecture Topic
21st Sept
wk1
 Introduction to Data Mining, Machine Learning and Data Science
28th Sept
wk2
 Data Mining / Data Science Life-cycle, Why it is different and the challenges
5th Oct
wk3
 Descriptive Analytics and Preparing Data
12th Oct
wk4
Exploring Text & Basics of Text Mining  follow videos and exercises.
19th Oct
wk5
Association Rule Analysis

Important links for Product Support

SAS Community Forum

SAS Enterprise Miner Documentation

26th Oct
wk6
Bank Holiday – Catchup week. Complete & Repeat Exercises.

There is No lectures this week.  Instead use this week to catch-up on all the lab exercises, repeating them to get a better understanding, and loading new data into the tool to explore.

Source a Data Set from one of the many data set repositories, load it into SAS EM and use the tool to explore the data. See notes from previous weeks on how to process and explore this data.

This link has a list of some data set repositories.

2nd Nov
wk7
 Classification
9th Nov
wk8
Classification

 Assignment Handout – Groups of max. 2 people or individual.

16th Nov
wk9
 Finish Classification – see previous weeks notes

Data Clusteringwatch videos and lab exercises

I’ll be available for first 90 minutes of scheduled class time for a Q&A on topic for this week and to answer any questions about the Assignment.

23rd Nov
wk10
Deployment of Data Science, Data Mining, Machine Learning and Artificial Intelligence Projects

Explore the Deployment node of CRISP-DM and the outer ring. MLOps/AIOps will be covered, along with other deployment considerations for your DS/DM/ML projects

30th Nov
wk11
EU GDPR, Laws and How Ethics affects Data Science, Data Mining, Machine Learning and Artificial Intelligence Projects

Watch the videos and lab exercises. Use remaining time to work on Assignment.

I’ll be available for first 40 minutes of scheduled class time for a Q&A on topic for this week and to answer any questions about the Assignment.

7th Dec
wk12
Work on Assignment – Assignment Due Wednesday 16th December @23:00
No Class – Use the class and lab time to work on completing your assignment.
14th Dec
wk13
Work on Assignment – Assignment Due Wednesday 16th December @23:00
No Class – complete and submit assignment.
Jan IMPORTANT : Some minor updates highlighted in Red. I’ve just been made aware of these updates by management in TU Dublin.

Exam will be an Open Book Offline exam.   It looks like the exam is scheduled for 12th January between 09:00-12:00 (correction to time – it will be a 3 hour exam).

The Exam Paper will be released/available on Brightspace  (@9am on 12th January)

You will have 3 hours to complete the paper and submit your work before 12:00 (No extensions)

The exam paper will have 4 questions. You need to complete 3 questions.

Question 1 has 34 marks and all other questions have 33 marks.

Use a word processor to complete all your answers, include all diagrams, examples, etc (everything) in this document. Convert to PDF and upload this file before the deadline (12:00) in Brightspace.

Locked Diagrams in SAS Enterprise Miner

*** The following is from SAS

In regards to the locked diagram, the fastest way to delete the lock typically is to log into SAS OnDemand for Academics Control Center. When a diagram is opened for editing, a lock file is put in place to indicate the diagram is being used. Sometimes, the lock file will remain even when the user is finished editing the diagram. As a result, the user will get an error in a pop-up window stating that “this diagram is currently locked”. To correct this problem, you should close the diagram if it is open. Then open SAS Studio from the SAS OnDemand for Academics Control Center. Open the “Server Files and Folders” tab. The lock file will generally be in this path:

 /home/(user name)/(project name)/Workspaces/EMWS1 (or similar)/System

 ****The “.lck” file I deleted was found at the following filepath in your Enterprise Miner project files in SAS Studio: /home/basquinsophonie0/Exam1/Workspaces/EMWS1/System

 If your diagram locks up on you again, you can head to that file location and delete the “.lck” file, which should allow you to open the diagram.****

 The lock file will have a name like “wsopen.lck”. (The lock file always has the extension “.lck”. )

 You can remove this file to enable editing of the diagram.