Identification of statistical patterns in complex systems via symbolic time series analysis

Gupta, Savita; Khatkhate, Amol M; Ray, Asok; Keller, Eric

doi:10.1016/s0019-0578(07)60226-2

Cited by 6 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More techniques on Arena software can refer to Altiok and Melamed. 21 For empirical study of traffic in a transportation system, Zhang 22 use matrix geometric method to model the traffic flow at tollbooth plaza in China, and Gupta et al 23 perform identification of statistical patterns in complex systems via symbolic time-series analysis. The study on Yangtze River waterway is growing such as Zhang and Zhao 1 and Liu.…”

Section: Literature Reviewmentioning

confidence: 99%

Simulation modeling of state-dependent queuing network: Impact of deepening on the vessel traffic in Yangtze River Estuary

Hao

Dong

et al. 2019

Advances in Mechanical Engineering

View full text Add to dashboard Cite

This article is to study the impact of deepening on transportation efficiency in Yangtze River Estuary. Due to geographic conditions and administrative regulations, navigation system at the estuary can be formulated as a state-dependent queuing network with interdependent channels, multi-type vessels, and priority disciplines. We use simulation modeling to deal with this complex maritime system. Vessel calls to terminals, tidal and navigational rules, and vessel profiles are considered and a ''dual clock'' mechanism is designed to solve the pre-queuing optimization. The model is specifically built to stimulate vessel traffic on performance measures, including vessel waiting time and channel utilization. By performing scenario and policy analyses, several proposed deepening schemes are evaluated and compared.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Simulation modeling of state-dependent queuing network: Impact of deepening on the vessel traffic in Yangtze River Estuary

Hao

Dong

et al. 2019

Advances in Mechanical Engineering

View full text Add to dashboard Cite

show abstract

“…Entropy-based anomaly detection approaches have gained substantial attention in the research community during last few years (He et al, 2008; Sharif and Djeraba, 2012). The appealing feature of entropy lies in the fact that entropy is capable of reducing the entire feature distribution into a single number which retains the critical information about the overall state of distribution (Cammarota and Rogora, 2005; Gupta et al, 2006a; Ishizaki and Inoue, 2013; Rajagopalan et al, 2007). In this context, estimation of deviation of a system’s behavior from its nominal behavior can be realized by monitoring the entropy of the system.…”

Section: Proposed Damage Identification Strategymentioning

confidence: 99%

Structural condition assessment using entropy-based time series analysis

Alamdari

Samali

et al. 2017

Journal of Intelligent Material Systems and Structures

View full text Add to dashboard Cite

We present a time-series-based algorithm to identify structural damage in the structure. The method is in the context of non-model-based approaches; hence, it eliminates the need of any representative numerical model of the structure to be built. The method starts by partitioning the state space into a finite number of subsets which are mutually exclusive and exhaustive and each subset is identified by a distinct symbol. Partitioning is performed based on a maximum entropy approach which takes into account the sparsity and distribution of information in the time series. After constructing the symbol space, the time series data are uniquely transformed from the state space into the constructed symbol space to create the symbol sequences. Symbol sequences are the simplified abstractions of the complex system and describe the evolution of the system. Each symbol sequence is statistically characterized by its entropy which is obtained based on the probability of occurrence of the symbols in the sequence. As a consequence of damage occurrence, the entropy of the symbol sequences changes; this change is implemented to define a damage indicative feature. The method shows promising results using data from two experimental case studies subject to varying excitation. The first specimen is a reinforced concrete jack arch which replicates one of the major structural components of the Sydney Harbor Bridge and the second specimen is a three-story frame structure model which has been tested at Los Alamos National Laboratory. The method not only could successfully identify the presence of damage but also has potential to localize it.

show abstract

“…The process involves defining a mapping that translates a given data sequence () to a symbolic space with , where is the length of the enclosed vector. This mapping can be achieved in many different ways including representing in a multidimensional state-space [28], [33]–[35]. Once symbolized the sequence is either clustered as patterns (such as words from character sequences) or a measure is applied to quantify symbolic dynamics, the latter typically employed using entropy-based measures [30], [36].…”

Section: Symbolic Analysis Of Eeg Signalsmentioning

confidence: 99%

Ngram-Derived Pattern Recognition for the Detection and Prediction of Epileptic Seizures

et al. 2014

View full text Add to dashboard Cite

This work presents a new method that combines symbol dynamics methodologies with an Ngram algorithm for the detection and prediction of epileptic seizures. The presented approach specifically applies Ngram-based pattern recognition, after data pre-processing, with similarity metrics, including the Hamming distance and Needlman-Wunsch algorithm, for identifying unique patterns within epochs of time. Pattern counts within each epoch are used as measures to determine seizure detection and prediction markers. Using 623 hours of intracranial electrocorticogram recordings from 21 patients containing a total of 87 seizures, the sensitivity and false prediction/detection rates of this method are quantified. Results are quantified using individual seizures within each case for training of thresholds and prediction time windows. The statistical significance of the predictive power is further investigated. We show that the method presented herein, has significant predictive power in up to 100% of temporal lobe cases, with sensitivities of up to 70–100% and low false predictions (dependant on training procedure). The cases of highest false predictions are found in the frontal origin with 0.31–0.61 false predictions per hour and with significance in 18 out of 21 cases. On average, a prediction sensitivity of 93.81% and false prediction rate of approximately 0.06 false predictions per hour are achieved in the best case scenario. This compares to previous work utilising the same data set that has shown sensitivities of up to 40–50% for a false prediction rate of less than 0.15/hour.

show abstract

Identification of statistical patterns in complex systems via symbolic time series analysis

Cited by 6 publications

References 20 publications

Simulation modeling of state-dependent queuing network: Impact of deepening on the vessel traffic in Yangtze River Estuary

Simulation modeling of state-dependent queuing network: Impact of deepening on the vessel traffic in Yangtze River Estuary

Structural condition assessment using entropy-based time series analysis

Ngram-Derived Pattern Recognition for the Detection and Prediction of Epileptic Seizures

Contact Info

Product

Resources

About