A Scalable Data Analytics Algorithm for Mining Frequent Patterns from Uncertain Data

With advances in technology, massive amounts of valuable data can be collected and transmitted at high velocity in various scientific, biomedical or engineering applications. Hence, scalable data analytics tools are in demand for analyzing these data. For example, scalable tools for association analysis help reveal frequently occurring patterns and their relationships, which in turn lead to intelligent decisions. While a majority of existing frequent pattern mining algorithms—including FPgrowth—handle only precise data, there are situations in which data are uncertain. In recent years, researchers have paid attention to frequent pattern mining from uncertain data. UF-growth and UFP-growth are examples of tree-based algorithms for mining uncertain data. However, their corresponding tree structures can be large. Other tree structures for handling uncertain data may achieve compactness at the expense of loose upper bounds on expected supports. To solve this problem, we propose (i) a compact tree structure that captures uncertain data with tighter upper bounds than aforementioned tree structures and (ii) a scalable data analytics algorithm that mines frequent patterns from our tree structure. Experimental results show the tightness of bounds to expected supports provided by our algorithm.