A fuzzy/Bayesian approach for the time series change point detection problem

D’Angelo, Marcos Flávio Silveira Vasconcelos; Palhares, Reinaldo M.; Takahashi, Ricardo H. C.; Loschi, Rosangela H.

doi:10.1590/s0101-74382011000200002

Cited by 3 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the disadvantages include difficulty in calculating density estimation for high-dimensional data, which can be slow and less robust. The authors in [19] have proposed a fuzzy Bayesian change-point detection technique using the posterior probability of the current run length in time-series data. The proposed technique works in two folds.…”

Section: Introductionmentioning

confidence: 99%

Optimal Parameter Exploration for Online Change-Point Detection in Activity Monitoring Using Genetic Algorithms

Khan

McClean

Zhang

et al. 2016

Sensors

View full text Add to dashboard Cite

In recent years, smart phones with inbuilt sensors have become popular devices to facilitate activity recognition. The sensors capture a large amount of data, containing meaningful events, in a short period of time. The change points in this data are used to specify transitions to distinct events and can be used in various scenarios such as identifying change in a patient’s vital signs in the medical domain or requesting activity labels for generating real-world labeled activity datasets. Our work focuses on change-point detection to identify a transition from one activity to another. Within this paper, we extend our previous work on multivariate exponentially weighted moving average (MEWMA) algorithm by using a genetic algorithm (GA) to identify the optimal set of parameters for online change-point detection. The proposed technique finds the maximum accuracy and F_measure by optimizing the different parameters of the MEWMA, which subsequently identifies the exact location of the change point from an existing activity to a new one. Optimal parameter selection facilitates an algorithm to detect accurate change points and minimize false alarms. Results have been evaluated based on two real datasets of accelerometer data collected from a set of different activities from two users, with a high degree of accuracy from 99.4% to 99.8% and F_measure of up to 66.7%.

show abstract

Section: Introductionmentioning

confidence: 99%

Optimal Parameter Exploration for Online Change-Point Detection in Activity Monitoring Using Genetic Algorithms

Khan

McClean

Zhang

et al. 2016

Sensors

View full text Add to dashboard Cite

show abstract

Monitoring the Stator Current in Induction Machines for Possible Fault Detection: A Fuzzy/Bayesian Approach for the Problem of Time Series Multiple Change Point Detection

et al. 2016

Self Cite

View full text Add to dashboard Cite

ABSTRACT. This paper addresses the problem of fault detection in stator winding of induction machine by a multiple change points detection approach in time series. To handle this problem a new fuzzy/Bayesian approach is proposed which differs from previous approaches since it does not require prior information as: the number of change points or the characterization of the data probabilistic distribution. The approach has been applied in the monitoring the current of the stator winding induction machine. The good results obtained by proposed methodology illustrate its efficiency.

show abstract

Fuzzy Inference Systems for Multi-Step Ahead Daily Inflow Forecasting

et al. 2017

View full text Add to dashboard Cite

This paper presents the evaluation of a daily inflow forecasting model using a tool that facilitates the analysis of mathematical models for hydroelectric plants. The model is based on a Fuzzy Inference System. An offline version of the Expectation Maximization algorithm is employed to adjust the model parameters. The tool integrates different inflow forecasting models into a single physical structure. It makes uniform and streamlines the management of data, prediction studies, and presentation of results. A case study is carried out using data from three Brazilian hydroelectric plants of the Parana basin, Tiete River, in southern Brazil. Their activities are coordinated by Operator of the National Electric System (ONS) and inspected by the National Agency for Electricity (ANEEL). The model is evaluated considering a multi-step ahead forecasting task. The graphs allow a comparison between observed and forecasted inflows. For statistical analysis, it is used the mean absolute percentage error, the root mean square error, the mean absolute error, and the mass curve coefficient. The results show an adequate performance of the model, leading to a promising alternative for daily inflow forecasting.

show abstract

A fuzzy/Bayesian approach for the time series change point detection problem

Cited by 3 publications

References 14 publications

Optimal Parameter Exploration for Online Change-Point Detection in Activity Monitoring Using Genetic Algorithms

Optimal Parameter Exploration for Online Change-Point Detection in Activity Monitoring Using Genetic Algorithms

Monitoring the Stator Current in Induction Machines for Possible Fault Detection: A Fuzzy/Bayesian Approach for the Problem of Time Series Multiple Change Point Detection

Fuzzy Inference Systems for Multi-Step Ahead Daily Inflow Forecasting

Contact Info

Product

Resources

About