Abstract:Along with the development of monitoring technologies, numerous measured data pour into monitoring system and form the high-volume and open-ended data stream. Usually, abnormal condition of monitored system can be characterized by the density variation of measured data stream. However, traditional density estimation methods can not dynamically track density variation of data stream due to the limitation of processing time and computation memory. In this paper, we propose a new density estimation neural network to continuously estimate the density of streaming data in a time-based sliding window. The network has a feedforward structure composed of discretization, input and summation layer. In the discretization layer, value range of data stream is discretized to network nodes with equal intervals. Measured data in the predefined time window are pushed into input layer and updated with the window sliding. In summation layer, the activation results between input neurons and discretization neurons are summed up and multiplied by a weight factor. The network outputs the kernel density estimators of sliding segment in data stream and achieves a one-pass estimation algorithm consuming constant computation memory. By subnet separation and local activation, computation load of the network is significantly reduced to catch up the pace of data stream. The nonlinear statistics, quantile and entropy, which can be consecutively figured out with the density estimators output by the density estimation neural network, are calculated as condition indictors to track the density variation of data stream. The proposed method is evaluated by a simulated data stream consisting of two mixing distribution data sets and a pressure data stream measured from a centrifugal compressor respectively. Results show that the underlying anomalies are successfully detected.

Density Estimation Over Data Stream ∗

M-kernel Merging: Towards Density Estimation over Data Streams

Fast Rotation Kernel Density Estimation over Data Streams

Density Estimation Based on Mass

Density-Based Clustering over an Evolving Data Stream with Noise

A New Density Estimation Neural Network to Detect Abnormal Condition in Streaming Data

Data Stream Clustering by Fast Density-Peak-search

Demass: A New Density Estimator for Big Data

A Clustering Algorithm Based on Density-Grid for Stream Data

Efficient Computation of K-Medians over Data Streams under Memory Constraints

Clustering Stream Data by Exploring the Evolution of Density Mountain

Dynamic Density-Based Clustering Algorithm over Uncertain Data Streams

Granularity adaptive density estimation and on demand clustering of concept-drifting data streams

Efficient Data Stream Clustering Algorithm Based on k-Means Partitioning and Density

Density-based clustering of data streams at multiple resolutions

Continuous Adaptive Outlier Detection on Distributed Data Streams.

Density-based hierarchical clustering for streaming data

Distribution-free Data Density Estimation in Large-Scale Networks

Summarization and Matching of Density-Based Clusters in Streaming Environments

A Generic Sketch for Estimating Super-Spreaders and Per-Flow Cardinality Distribution in High-Speed Data Streams

An Adaptive Density Data Stream Clustering Algorithm