MBA504 Workshop Nine Demonstration of BI Tools Data Analysis, Problem Solving and Digital OperationsCOMMONWEALTH OF AUSTRALIA Copyright Regulations 1969 WARNING This material has been reproduced and communicated to you by or on behalf of Kaplan Business School pursuant to Part VB of the Copyright Act 1968 (the Act). The material in this communication may be subject to copyright under the Act. Any further reproduction or communication of this material by you may be the subject of copyright protection under the Act. Do not remove this notice. 2USING BI TOOLS â€¢ We will explore some basic operations with common BI tools â€“ Tableau and Spotfire.EXAMPLE OF A TYPICAL BI TOOL Connect to Data The first thing to do when you start is to connect to your data. For example in Tableau: 4Build a View â€“ After you connect to a data source, fields are displayed in the Data pane on the left side of the workbook as dimensions and measures. 6 EXAMPLEACTIVITY The preceding examples used screen shots from Tableau. Try to recreate the steps with a trial version of your BI tool of choice.SOLUTION Activity: Whilst the diagram on this slide illustrates the solution, only you are able to articulately interpret it. In groups, develop a 2-minute presentation that explains this analysis in an easy-to-understand manner.STOCK SORTING This data pertains to stocks collected over a six month period. The goal is to use K-means clustering (a machine learning technique) to organise stocks based on the trend in their closing price.TIME SERIES OF STOCK PRICES The line chart shows the trend of stocks. We will use K-means clustering to organise these stocks by trend. The magnitude of the stock prices are not important.K-MEANS CLUSTERING â€œK-means clustering is an unsupervised machine learning algorithm for partitioning a dataset into separate clusters where instances (of the dataset) in each cluster are similar to each other.â€ (Livani, Nguyen, Denzinger, Ruhe, & Banack 2013, p. 169)K-MEANS CLUSTERING Activity: In groups, discuss the following questions: â€¢ How many clusters is a reasonable number to have? â€¢ What are the strengths and limitations of the method? â€¢ What is the general interpretation/idea/usefulness of each cluster? â€¢ Is there only one â€œoptimal clusterâ€ or are there sometimes non-unique, and still useful, ways of forming clusters?