A Rapid Incremental Frequent Pattern Mining Algorithm for Uncertain Data
Tu‐Liang Lin, Bo-Wei Wen, Hong‐Yi Chang, Wan-Kun Chang, Shih-Che Hsu · 2017
Association rule analysis is an important topic in data mining. Basket analysis is one of the most well-known applications. Store or retailer can get better sales through the analysis of goods combination. For example, placing beer and diapers at the same place can bring greater sales for the store. However, due to the rapid increase in the amount of data in this big data era, how to mine frequent patterns from big data has become an important issue. Many approaches were proposed to solve the incremental problem of certain data, but these approaches did not address uncertain data. The CUF-Growth algorithm preventing branches improves the performance of the traditional UF-Growth. In this paper, we propose an incremental association algorithm based on CUF-Growth to solve the problem of incremental updating of uncertain frequent items. This method retains the advantages of the original CUF-Growth, and significantly reduces the complexity of adding new transactions. The experimental results show that the proposed method reduces the execution time and perform better than the traditional UF-Growth.