Abstract:Pattern discovery is a widely discussed topic in the fields of data mining and artificial intelligence. It involves the complex process of uncovering meaningful and potentially valuable information from a vast amount of data stored in a database. This process, commonly known as pattern mining, requires advanced techniques and algorithms to extract insights that were previously unknown or implicit in the data. In recent years, periodic pattern mining (PPM) has become a hot topic. Traditional PPM assumes that the periodicity of the pattern has better stability within a certain time interval. To address this limitation, local periodic pattern (LPP) mining, where the pattern appears continuously within a certain time interval, is proposed. However, in practical applications, not all LPPs are potentially valuable and interesting. Therefore, targeted mining based on user preferences has become urgent. This paper presents the problem of target-based LPPM for the first time and elaborates on the task of mining targeted local periodic patterns (TaLPPs) in a temporal transaction database. A depth-first algorithm based on a vertical list structure is proposed, named TaLPPM, which aims to mine TaLPPs in a temporal transaction database accurately. Using a new pattern-matching mechanism, TaLPPM can quickly match target patterns in the process of mining LPPs. TaLPPM proves the downward closure property of the target pattern and uses it to quickly determine whether an itemset matches the target pattern. In addition, TaLPPM introduces a pruning strategy based on the characteristics of the target pattern to reduce the search space. Extensive experiments on several real and synthetic datasets show that TaLPPM is satisfactory in both correctness and mining performance.

Discovering Periodic Patterns Common to Multiple Sequences

Stable Periodic Frequent Itemset Mining on Uncertain Datasets

On Discovering Feasible Periodic Patterns in Large Database.

Fuzzy-driven periodic frequent pattern mining

Discovering Local Patterns From Multiple Temporal Sequences

Efficient Discovery of Periodic Patterns over Event Sequences

Utility-driven Mining of Contiguous Sequences

An Asynchronous Periodic Sequential Patterns Mining Algorithm with Multiple Minimum Item Supports.

Mining Asynchronous Periodic Patterns in Time Series Data.

TSPIN: mining top-k stable periodic patterns

Mining Follow-Up Correlation Patterns from Time-Related Databases

Efficient Mining Of Short Periodic High-Utility Itemsets

Mining Sequential Patterns with Periodic General Gap Constraints

Mining Periodic High-Utility Itemsets with Both Positive and Negative Utilities.

Mining Weighted Sequential Patterns In A Sequence Database With Itemset-Interval Measurement

Mining Frequent Itemsets Over Arbitrary Time Intervals in Data Streams

Mining multi-level time-interval sequential patterns in sequence databases

Targeted Mining of Local Periodic Patterns

Mining Maximal Sequential Patterns

Mining frequent temporal duration-based patterns on time interval sequential database

Discovering Frequent Patterns in Sequence Groups