An efficient approach to mine flexible periodic patterns in time series databases

Chanda, Ashis Kumar; Saha, Swapnil; Nishi, Manziba Akanda; Samiullah, M.; Ahmed, Chowdhury Farhan

doi:10.1016/j.engappai.2015.04.014

Cited by 30 publications

(16 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recent data mining techniques find “partial periodic patterns” that occur repeatedly in time series, sometimes with imperfect regularity [ 5 , 6 , 12 , 25 , 27 – 29 , 32 , 34 , 39 , 40 , 61 , 62 ]. These methods represent discrete time series as strings and search for recurring patterns.…”

Section: Related Workmentioning

confidence: 99%

Modeling Individual Cyclic Variation in Human Behavior

Pierson

Althoff

Leskovec

2018

Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18

View full text Add to dashboard Cite

Cycles are fundamental to human health and behavior. Examples include mood cycles, circadian rhythms, and the menstrual cycle. However, modeling cycles in time series data is challenging because in most cases the cycles are not labeled or directly observed and need to be inferred from multidimensional measurements taken over time. Here, we present Cyclic Hidden Markov Models (CyH-MMs) for detecting and modeling cycles in a collection of multidimensional heterogeneous time series data. In contrast to previous cycle modeling methods, CyHMMs deal with a number of challenges encountered in modeling real-world cycles: they can model multivariate data with both discrete and continuous dimensions; they explicitly model and are robust to missing data; and they can share information across individuals to accommodate variation both within and between individual time series. Experiments on synthetic and real-world health-tracking data demonstrate that CyHMMs infer cycle lengths more accurately than existing methods, with 58% lower error on simulated data and 63% lower error on real-world data compared to the best-performing baseline. CyHMMs can also perform functions which baselines cannot: they can model the progression of individual features/symptoms over the course of the cycle, identify the most variable features, and cluster individual time series into groups with distinct characteristics. Applying CyHMMs to two real-world health-tracking datasets—of human menstrual cycle symptoms and physical activity tracking data—yields important insights including which symptoms to expect at each point during the cycle. We also find that people fall into several groups with distinct cycle patterns, and that these groups differ along dimensions not provided to the model. For example, by modeling missing data in the menstrual cycles dataset, we are able to discover a medically relevant group of birth control users even though information on birth control is not given to the model.

show abstract

Section: Related Workmentioning

confidence: 99%

Modeling Individual Cyclic Variation in Human Behavior

Pierson

Althoff

Leskovec

2018

Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18

View full text Add to dashboard Cite

show abstract

“…Where I ij = { 1, I j ∈ T i 0, I j ∉ T i ，i = 1,2, … n，j = 1,2, … m，C(T i ) represents the times transaction T i appears in the transaction database D. By scanning the original transaction database, transactions are de-duplicated and counted, then are stored in matrix M. If there are many duplicate transactions in the transaction database, the original transaction database can be greatly compressed [6,7] .…”

Section: Definitionsmentioning

confidence: 99%

A Running Pattern Recognition Algorithm for Publish/Subscribe Distributed Systems

Wu¹,

Shen²,

Wang³

2021

dtmse

View full text Add to dashboard Cite

Publish/subscribe distributed systems are often used in critical applications. It is necessary to monitor their running patterns in real time to detect abnormal status. Therefore, identifying the normal running pattern is the precondition of monitoring publish/subscribe distributed systems. Based on Apriori algorithm, this paper presents a weighted frequent itemset mining algorithm for running pattern recognition of publish/subscribe distributed systems. By introducing the transaction matrix, the algorithm only needs to scan the transaction database once. By weighting the items from two aspects of influence and frequency, the support of the items with few occurrences but much importance can be improved, so that the running pattern containing small frequency events can be mined out. Experimental results show that the algorithm can effectively mine the running patterns, and has better performance than Apriori algorithm and FP-growth algorithm.

show abstract

“…A pattern is known as periodic when it reoccurs in a time-series database within a fixed time interval. Several useful definitions used in this paper are given below, for a more thorough treatment of pattern mining and periodicity detection we refer to the work of Nishi et al and Chanda et al [12,13].…”

Section: Preliminariesmentioning

confidence: 99%