Discovering high utility-occupancy patterns from uncertain data

It is widely known that there a lot of useful information hidden in big data, leading to new saying "data money." Thus, it prevalent for individuals mine crucial utilization many real-world applications. In the past, studies have considered frequency. Unfortunately, doing so neglects other aspects, such as utility, interest, or risk. sensible discover high-utility itemsets (HUIs) transaction databases while utilizing not only quantity but also predefined utility. To find patterns can represent supporting transaction, recent study was conducted high utility-occupancy whose contribution utility entire greater than certain value. Moreover, realistic applications, may exist transactions be connected an existence probability. this paper, novel algorithm, called High-Utility-Occupancy Pattern Mining Uncertain (UHUOPM), proposed. The found by algorithm are Potential High Utility Occupancy Patterns (PHUOPs). This divides user preferences into three factors, including support, probability, and occupancy. reduce memory cost time consumption prune search space mentioned above, probability-utility-occupancy list (PUO-list) probability-frequency-utility table (PFU-table) used, which assist providing downward closure property. Furthermore, original tree structure, support count (SC-tree), constructed algorithm. Finally, substantial experiments were evaluate performance proposed UHUOPM on both real-life synthetic datasets, particularly terms effectiveness efficiency.