A unifying framework for detecting outliers and change points from non-stationary time series data

By Kenji Yamanishi, Jun'ichi Takeuchi. Knowledge Discovery and Data Mining, 2002

352 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/1885b13d0e3e478d4f2034d3d469fba689543ef4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Deep Learning for Time Series Anomaly Detection: A Survey (2022, neutral): “According to [77], the term anomaly refers to a deviation from the general distribution of data, such as a single observation (point) or a series of observations (subsequence) that deviate greatly from the general distribution.”
A review on time series data mining (2011, background): “Yamanishi and Takeuchi (2002) present an online learning framework based on a probabilistic model for outlier detection and change-point detection on the time series data stream.”
Incremental Local Outlier Detection for Data Streams (2007, background): “In contrast, incremental outlier detection techniques [38, 39, 41] identify outliers as soon as new data record appears in the dataset.”
Change-Point Detection in Time-Series Data by Relative Density-Ratio Estimation (2012, methodology): “…datasets that contain manually inserted change-points: • Dataset 1 (Jumping mean): The following 1-dimensional auto-regressive model borrowed from Takeuchi and Yamanishi (2006) is used to generate 5000 samples (i.e., t = 1 , . . . , 5000): where ǫ t is a Gaussian noise with mean µ and…”
A Survey of Methods for Time Series Change Point Detection (2016, neutral): “Change Finder [2][45][22] is another commonly used method which reduces the problem of change point detection into time series-based outlier detection.”
Change-Point Detection in Time-Series Data by Direct Density-Ratio Estimation (2009, methodology): “Change finder (CF) [33, 34]: CF first sequentially fits an AR model to time-series data and auxiliary time-series data is generated from the AR model.”
Outlier Detection for Temporal Data: A Survey (2014, methodology): “[102], [103] present SmartSifter which employs an online discounting learning algorithm to incrementally learn the probabilistic mixture model, with a decay factor to account for drift.”
Oboe: auto-tuning video ABR algorithms to network conditions (2018, methodology): “While several techniques exist for change point detection [22, 33, 36, 44, 54, 58], we focus on probabilistic methods [14, 18, 24, 57].”