An Iterative Method for Mining Frequent Temporal Patterns

The incorporation of temporal semantic into the traditional data mining techniques has caused the creation of a new area called Temporal Data Mining. This incorporation is especially necessary if we want to extract useful knowledge from dynamic domains, which are timevarying in nature. However, this process is computationally complex, and therefore it poses more challenges on efficient processing that nontemporal techniques. Based in the inter-transactional framework, in [11] we proposed an algorithm named TSET for mining temporal patterns (sequences) from datasets which uses a unique tree-based structure for storing all frequent patterns discovered in the mining process. However, in each data mining process, the algorithm must generate the whole structure from scratch. In this work, we propose an extension which consists in the reusing of structures generated in previous data mining process in order to reduce the execution time of the algorithm.

• Type: Conference
• Year: 2005
• Where: EUROCAST
• Authors: Francisco Guil, Antonio B. Bailón, Alfonso Bosch, Roque Marín
• Keywords: Data Mining, Data Mining Process, EUROCAST 2005, Mining Process