Comparing GUHA and Weka Methods in Data Mining
Permanent address of the item is
The development of computers has enabled the collection and storage of terabytes of data and the creation of large data warehouses. The main problems with such data are their size and structure. The fundamental intellectual challenges at present are the analysis and understanding of the data in the decision-making process. This thesis introduces and compares the methods of GUHA and Weka software. The thesis highlights the differences between GUHA and Weka software through taking 2 methods which reveal the association rules of the Weka programme and comparing them with three methods which reveal the association rules of the GUHA programme. The difficulty of the task is the amount of computation which has to be done to explain whether the methods have any differences or not. The work has been done by taking the results from one of the Weka methods and comparing these with all the methods of GUHA. The second Weka method provides the same results as the first one, but in a different order. The results have been carefully compared and there are some comments in the discussion part of the thesis.