The objective of the study focused on weighted based frequent item set mining. The base paper has proposed multi criteria based frequent item set for weight calculation. Contribution towards this project is to implement the global profit weight measure and test the performance over utility based mining. For this project the data consist of 90 products from automobile shop including unit price, quantity sold and profit margin for transaction set (one month data). Algorithm has been implemented in Visual Basic for visualizing step by step process calculations. Supervised machine learning techniques namely Na?ve Bayes Decision tree classifier, VFI and IB1 Classifier are used for learning the model. The results of the models are compared and observed that Na?ve Bayes performs well. WEKA tool is used to classify the data set and accuracy is calculated.
展开▼
