An Information-Theoretic Approach to Detecting Changes in Multi-Dimensional Data Streams
Explore this paper's citation graph
Summary
This paper uses relative entropy, also called the Kullback-Leibler distance, to measure the difference between two given distributions, which generalizes Kulldorff’s spatial scan statistic, allowing us to quantitatively identify specific regions in space where large changes have occurred.
- Type
- article
- Published
- 2006-01-01
- Cited by
- 261
- References
- 34
- OpenAlex
- https://openalex.org/W3487859
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:10591316
Keywords
STREAMS, Computer science, Data mining, Data stream mining
References
- Statistical Decision Rules and Optimal Inference
- Information Theory: 1948-1998 - Guest Editorial
- Bursty and Hierarchical Structure in Streams
- Logistic Regression, AdaBoost and Bregman Distances
- Mining Surprising Patterns Using Temporal Description Length
- Exploratory Data Mining and Data Cleaning
- Representing and querying changes in semistructured data
- Efficient elastic burst detection in data streams
- A Framework for Measuring Differences in Data Characteristics
- Finding surprising patterns in a time series database in linear time and space
- The performance of universal encoding
- A framework for diagnosing changes in evolving data streams
- Mining time-changing data streams
- Asymptotically optimal classification for multiple tests with empirically observed statistics
- Generalized Linear Models
- Exponentiated Gradient Versus Gradient Descent for Linear Predictors
- Rapid detection of significant spatial clusters
- Handbook of Parametric and Nonparametric Statistical Procedures
- A Divisive Information-Theoretic Feature Clustering Algorithm for Text Classification
- Meaningful change detection in structured data
Cited by
- Black Box Anomaly Detection: Is It Utopian?
- Managing and Mining Sensor Data
- Non-parametric Information-Theoretic Measures of One-Dimensional Distribution Functions from Continuous Time Series
- Nuevos métodos para el aprendizaje en flujos de datos no estacionarios
- Streaming techniques for statistical modeling
- Mining Time-Changing Data Streams
- Change Detection in Learning Histograms from Data Streams
- Non-Parametric Methods Applied to the N-Sample Series Comparison
- Classifying evolving data streams with partially labeled data
- Distance Functions to Detect Changes in Data Streams
- Multi-dimensional classification using Bayesian networks for stationary and evolving streaming data
- Detecting concept drift: An information entropy based method using an adaptive sliding window
- Data Streams: Models and Algorithms (Advances in Database Systems)
- Dynamic opponent modelling in two-player games
- Głębia położenia-rozrzutu w strumieniowej analizie danych ekonomicznych
- Fast and accurate detection of changes in data streams
- Extension et interrogation de résumés de flux de données. (Extending and querying data stream's summaries)
- A Statistical Description of Neural Ensemble Dynamics
- Frequent Patterns mining in time-sensitive Data Stream
- A simulation based approach to detect wear in industrial robots
Related papers
- Learning from Time-Changing Data with Adaptive Windowing
- Learning in the Presence of Concept Drift and Hidden Contexts
- A survey on concept drift adaptation
- Mining time-changing data streams
- Statistical change detection for multi-dimensional data
- A streaming ensemble algorithm (SEA) for large-scale classification
- CONTINUOUS INSPECTION SCHEMES
- Mining high-speed data streams