A new clustering method using wavelet based probability density functions for identifying patterns in time-series data

Kordestani, Mojtaba; Alkhateeb, Abedalrhman; Rezaeian, Iman; Rueda, Luis; Saif, Mehrdad

doi:10.1109/embsisc.2016.7508616

Cited by 12 publications

(7 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is calculated using a pyramid algorithm based on quadrature mirror filters convolutions [16]. Wavelets are fluctuations-like waves and oscillations contain centralized resources and characteristics in frequency and time domains [17], [18]. The orthogonal property and prior characteristics are more appropriate for tracking signals and fault diagnosis [19].…”

Section: A Discrete Wavelet Transformation (Dwt)mentioning

confidence: 99%

“…Malat introduced the DWT framework, known as (MRA) for planning any signal to altered decision-making levels [20]. So, suppose y (x) ∈ W 2 (S) where orthogonal wavelets and their scaling functions are displayed as linear combinations [17]: Such as ( ) and ( ) Specify the scaling function and orthogonal wavelets, respectively. Also n and k are the factors of dilation and translation.…”

Section: A Discrete Wavelet Transformation (Dwt)mentioning

confidence: 99%

See 1 more Smart Citation

Improved Technique to Diagnose Fault in IEEE Standard 14-Bus System

Amer*,

Selmy*,

Mohamed

2019

IJITEE

View full text Add to dashboard Cite

This study presents a new technique for fast detecting and diagnosing of power grids faults. Discrete Wavelet transform (DWT) has a major disadvantage of noise sensitivity. The proposed technique solves the problems of DWT, where a highprecision classification of noisy and faulty signals could be obtained. Fusion between voltage and power readings is done to provide a more reliable and accurate decision to determine the exact location of the fault. In this technique, the learner classifier is used,and the system is trained for multiple situations where most faults may occur. All simulations were carried out and performed on the standard IEEE 14 bus system to check the efficiency and performance of the technique proposed. Simulation results demonstrate, as will be discussed, a strong effectiveness of the suggested approach relative to others. The main feature of the proposed technique is that it can differentiate between faulty and noisy signals and recognize the fault's location quickly and with high reliability

show abstract

Section: A Discrete Wavelet Transformation (Dwt)mentioning

confidence: 99%

Section: A Discrete Wavelet Transformation (Dwt)mentioning

confidence: 99%

Improved Technique to Diagnose Fault in IEEE Standard 14-Bus System

Amer*,

Selmy*,

Mohamed

2019

IJITEE

View full text Add to dashboard Cite

show abstract

“…The simulated time series data contains 200 genes with 10 timepoints: 0, 2, 4, 6,8,18,24,32,48, and 72 mins. The data is composed of four distinct temporal patterns with 50 genes per pattern.…”

Section: Simulated Time Series Datamentioning

confidence: 99%

“…Many time series clustering algorithms have been introduced to understand the dynamics of biological processes. Clustering methods can be divided into two branches: hierarchical and partitioning [6]. Hierarchical clustering methods either build small clusters that iteratively merge into larger clusters or large clusters that divide into smaller clusters [1].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Lag Penalized Weighted Correlation for Time Series Clustering

Chandereng

Gitter

2018

Preprint

View full text Add to dashboard Cite

Motivation:The similarity or distance measure used for clustering can generate intuitive and interpretable clusters when it is tailored to the unique characteristics of the data. In time series datasets, measurements such as gene expression levels or protein phosphorylation intensities are collected sequentially over time, and the similarity score should capture this special temporal structure. Results: We propose a clustering similarity measure called Lag Penalized Weighted Correlation (LPWC) to group pairs of time series that exhibit closely-related behaviors over time, even if the timing is not perfectly synchronized. LPWC aligns pairs of time series profiles to identify common temporal patterns. It down-weights aligned profiles based on the length of the temporal lags that are introduced. We demonstrate the advantages of LPWC versus existing time series and general clustering algorithms. In a simulated dataset based on the biologically-motivated impulse model, LPWC is the only method to recover the true clusters for almost all simulated genes. LPWC also identifies distinct temporal patterns in our yeast osmotic stress response and axolotl limb regeneration case studies. Availability: The LPWC R package is available at https://github.com/gitter-lab/LPWC and CRAN under a MIT

show abstract