Periodic-frequent pattern mining is a critical technique for analyzing temporal data to identify recurring trends and patterns. However, traditional models face significant challenges, such as the rare item problem, where uniform frequency and periodicity assumptions can lead to either the exclusion of patterns involving rare items or the generation of excessive, trivial patterns. Additionally, these models often fail to capture patterns with partial periodicity, limiting their applicability in real-world scenarios where periodic behavior may be intermittent. To address these issues, advancements such as periodic-correlated pattern mining have been developed, incorporating measures like all-confidence and periodic-all-confidence to balance the significance of frequent and rare items. Furthermore, partial periodic pattern discovery models relax strict periodicity constraints, allowing for identifying patterns with intermittent periodic behavior. These innovations enhance the ability to extract valuable insights from complex temporal datasets, improving decision-making and strategic planning.