<![CDATA[The recording of State Property at Udayana University on SIMAK-BMN has been carried out well, but the recorded State Property data only supports the preparation of balance sheets, lists of goods, reports of goods and control cards. The amount of data that is increasing every year has not been utilized optimally. The large amount of data at Udayana University and the superiority of K-Means in data grouping makes it possible to group data at the preprocessing stage using K-Means, then forming decision tree rules using C5.0 which is a refinement of the ID3 and C4.5 algorithms. The variables used in the formation of the decision tree are conditions, useful life and warranty period. The results of the study show that the decision tree rules that are formed are first checking the useful life, secondly checking good and other than good conditions, thirdly checking damaged conditions, namely lightly damaged and heavily damaged, and fourthly checking the warranty period. The recommendations generated by the decision support system are 2,894 preserved items, 1,397 deleted items, 112 maintained items with guarantees and 18 maintained items without warranty, out of a total of 4,421 items. The accuracy obtained is 97.69% calculated using the confusion matrix. Keywords : State property; K-Means; C5.0]]>
Copyrights © 2023
