why arm is not good at mining numerical data – One common issue with most traditional Association Rule Mining, Singapore

 Why ARM is not good at mining numerical data 

Assignment Details:

(a) One common issue with most traditional Association Rule Mining (ARM) algorithms (e.g., the Apriori node) is their inability to mine numerical data without first converting them into categorical ones. Write a two-page research essay to discuss this issue and critically review at least two important research articles that attempt to address this issue. Preferably, at least one of these articles should discuss the use of binning. (40 marks)

(b) Identify a publicly available dataset for ARM. Evaluate and explain why this dataset can potentially produce useful association rules that are of value to data users. Provide details of dataset characteristics and its source (e.g., state the exact URL for download). A good source of candidate datasets can be found at this URL: https://archive.ics.uci.edu/ml/datasets.html). (10 marks)

(c) Construct an ARM model by applying an appropriate ARM node (available in the IBM SPSS Modeler) on the dataset stated in Part (b). The model details and interpretation of results should include the following:

(i) Compare the ARM node that you have chosen to use, with another ARM node that you have chosen not to use, in the IBM SPSS Modeler.

(ii) Report the parameters settings used and explain how you choose those settings;

(iii) The number of rules generated; and

(iv) Report at least two (2) interesting rules and their implications. (30 marks)