Mining Temporal Patterns in Time Interval-Based Data
Sequential pattern mining is an important subfield in knowledge mining. Recently, applications using time interval-based mostly event information have attracted considerable efforts in discovering patterns from events that persist for a few length. Since the link between two intervals is intrinsically complex, how to effectively and efficiently mine interval-based mostly sequences could be a challenging issue. During this paper, two novel representations, endpoint representation and endtime illustration, are proposed to simplify the processing of advanced relationships among event intervals. Based mostly on the proposed representations, three types of interval-based patterns: temporal pattern, incidence-probabilistic temporal pattern, and length-probabilistic temporal pattern, are outlined. In addition, we develop two novel algorithms, Temporal Pattern Miner (TPMiner) and Probabilistic Temporal Pattern Miner (P-TPMiner), to find three types of interval-based mostly sequential patterns. We tend to conjointly propose 3 pruning techniques to additional scale back the search area of the mining process. Experimental studies show that each algorithms are able to search out three types of patterns efficiently. Furthermore, we tend to apply proposed algorithms to real datasets to demonstrate the effectiveness and validate the practicability of proposed patterns.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here