logo

Data Mining Assignment Help

11 Pages1239 Words32 Views
   

Added on  2020-03-16

Data Mining Assignment Help

   Added on 2020-03-16

ShareRelated Documents
Data Mining[Pick the date]Student Id and Name
Data Mining Assignment Help_1
QUESTION 1 “Association rule” XLMiner generated output for the cosmetic data variables is furnished below:(Confidence % = 50%) (i)After examining the above output, the below highlighted comments can be made (Rouse,2004).Rule 1 If antecedent item brush has been acquired by the individual, and then there is 100 % associatedconfidence that consequent item nail polish will also be acquired by individual. 1
Data Mining Assignment Help_2
Rule 2 If antecedent item nail polish has been acquired by the individual, and then there is 63.22 %associated confidence that consequent item brushes will also be acquired by individual. Rule 3 If antecedent item nail polish has been acquired by the individual, and then there is 59.20 %associated confidence that consequent item bronzer will also be acquired by individual. (ii)The problem of rule redundancy is quite common is association rules. This essentiallyrefers to a situation when the support level of a given rule is quite accurately predicted bythe rule just preceding it. This can be illustrated by using Rule #2 as an example. Basedon the above output derived from XL miner, it is evident that corresponding support levelattached with each of these rules happens to be same. However, on a higher confidencelevel basis, rule 1 would be superior leading to rule 2 being termed as redundant (Ana,2014).For determining the utility of the rules, a key aspect is that they must be viewed incombination and not necessarily as isolated rules. However, on an individual basis, it iscritical to note that an association rule has two main aspects. One of these is the supportwhich is measured by lift ratio. The other aspect is confidence which is measured byconfidence level. Usually, if at a minimum one of these aspects tends to be high, the rulenormally would have practical utility and should not be ignored (Abramowics, 2013).(iii)The effect of change of confidence percentage would be analyzed here with the help ofXLMiner output. (Confidence % = 75%) 2
Data Mining Assignment Help_3
It can be observed from the list of rules of XLMiner output that when the minimum confidencehas shifted from 50% to 75%, then the number of rules has been significantly reduced. The set ofrules gets eliminated which has confidence % lesser than 75%. Also, the rule with 100%confidence % appears in the list of rule. Owing to the above understanding, it is recommendedthat this value should be kept at very high level as this could lead to elimination of certainrelations which might have significant support and hence should not be ideally discarded(Shumueli et.al., 2016).Question 2 (a)Dendrogram has been created that would be used to find the main clusters for the data set. 3
Data Mining Assignment Help_4

End of preview

Want to access all the pages? Upload your documents or become a member.

Related Documents
Business Case Analysis Data Mining
|10
|1041
|141

Data Mining | Assignment
|8
|771
|43

The Redundancy of Rule Tends
|10
|1164
|170

Assignment of Data Mining
|7
|706
|46

Business Case Analysis Association
|11
|1296
|45

Data Mining and Redundant Rule Assignment
|9
|993
|143