Reactive Soft Prototype Computing for Concept Drift Streams

Raab, Christoph; Heusinger, Moritz; Schleif, Frank-Michael

doi:10.1016/j.neucom.2019.11.111

Cited by 114 publications

(50 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…• KSWIN [18] is a window-based concept drift detection method that utilizes the Kolmogorov-Smirnov statistic test (KS-Test) to compare the distances of two distributions. This test is a nonparametric test that does not require any assumptions about the underlying data distribution.…”

Section: Concept Drift Detection Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Acoustic Scene Classifier Based on Gaussian Mixture Model in the Concept Drift Situation

Id¹,

Abe²,

Hara³

2021

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

The data distribution used in model training is assumed to be similar with that when the model is applied. However, in some applications, data distributions may change over time. This situation is called the concept drift, which might decrease the model performance because the model is trained and evaluated in different distributions. To solve this problem for scene audio classification, this study proposes the kernel density drift detection (KD3) algorithm to detect the concept drift and the combine-merge Gaussian mixture model (CMGMM) algorithm to adapt to the concept drift. The strength of the CMGMM algorithm is its ability to perform adaptation and continuously learn from stream data with a local replacement strategy that enables it to preserve previously learned knowledge and avoid catastrophic forgetting. KD3 plays an essential role in detecting the concept drift and supplying adaptation data to the CMGMM. Their performance is evaluated for four types of concept drift with three systematically generated scenarios. The CMGMM is evaluated with and without the concept drift detector. In summary, the combination of the CMGMM and KD3 outperforms two of four other combination methods and shows its best performance at a recurring concept drift.

show abstract

Section: Concept Drift Detection Methodsmentioning

confidence: 99%

“…• Active CMGMM adaptation: In this approach, the CMGMM actively detects the concept drift using a certain method and only adapts the model when the concept drift is detected. In this study, we compared KD3 to ADWIN [9], HDDMA, HDDMW [10], and KSWIN [18].…”

Section: Methodsmentioning

confidence: 99%

Acoustic Scene Classifier Based on Gaussian Mixture Model in the Concept Drift Situation

Id¹,

Abe²,

Hara³

2021

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

show abstract

“…1. For a comprehensive study of various CD types see Gama et al (2014) and Raab et al (2019). For this case, online classification algorithms use statistical tests to detect CD between two distributions.…”

Section: Stream Analysismentioning

confidence: 99%

Dimensionality reduction in the context of dynamic social media data streams

2021

Self Cite

View full text Add to dashboard Cite

In recent years social media became an important part of everyday life for many people. A big challenge of social media is, to find posts, that are interesting for the user. Many social networks like Twitter handle this problem with so-called hashtags. A user can label his own Tweet (post) with a hashtag, while other users can search for posts containing a specified hashtag. But what about finding posts which are not labeled by the creator? We provide a way of completing hashtags for unlabeled posts using classification on a novel real-world Twitter data stream. New posts will be created every second, thus this context fits perfectly for non-stationary data analysis. Our goal is to show, how labels (hashtags) of social media posts can be predicted by stream classifiers. In particular, we employ random projection (RP) as a preprocessing step in calculating streaming models. Also, we provide a novel real-world data set for streaming analysis called NSDQ with a comprehensive data description. We show that this dataset is a real challenge for state-of-the-art stream classifiers. While RP has been widely used and evaluated in stationary data analysis scenarios, non-stationary environments are not well analyzed. In this paper, we provide a use case of RP on real-world streaming data, especially on NSDQ dataset. We discuss why RP can be used in this scenario and how it can handle stream-specific situations like concept drift. We also provide experiments with RP on streaming data, using state-of-the-art stream classifiers like adaptive random forest and concept drift detectors. Additionally, we experimentally evaluate an online principal component analysis (PCA) approach in the same fashion as we do for RP. To obtain higher dimensional synthetic streams, we use random Fourier features (RFF) in an online manner which allows us, to increase the number of dimensions of low dimensional streams.

show abstract

“…It performs cuts in its window to better adapt to the learning algorithms. Kolmogorv-Smirnov Windowing (KSWIN) [25] follows a similar concept, but uses a more sensitive statistical test.…”

Section: Related Workmentioning

confidence: 99%

“…Tests summarize five runs of cross-validation on one million samples or complete real-world streams per runNeural Computing and Applicationsgive the impression that SAMKNN and OBA are superior to the LVQ variants, there are several advantages of the LVQ techniques. The experiments of[26] gives detailed comparison of the time and memory complexity of stream classifiers, clearly showing that RSLVQ is superior regarding time and memory complexity to SAMKNN and OBA. This makes the LVQ techniques more interesting for the real-time analysis of higher-dimensional data and analysis with less powerful devices like single-board computers.…”

mentioning

confidence: 99%

Passive concept drift handling via variations of learning vector quantization

Heusinger

Raab

Schleif

2020

Neural Comput & Applic

Self Cite

View full text Add to dashboard Cite

Concept drift is a change of the underlying data distribution which occurs especially with streaming data. Besides other challenges in the field of streaming data classification, concept drift has to be addressed to obtain reliable predictions. Robust Soft Learning Vector Quantization as well as Generalized Learning Vector Quantization has already shown good performance in traditional settings and is modified in this work to handle streaming data. Further, momentum-based stochastic gradient descent techniques are applied to tackle concept drift passively due to increased learning capabilities. The proposed work is tested against common benchmark algorithms and streaming data in the field and achieved promising results.

show abstract

Reactive Soft Prototype Computing for Concept Drift Streams

Cited by 114 publications

References 26 publications

Acoustic Scene Classifier Based on Gaussian Mixture Model in the Concept Drift Situation

Acoustic Scene Classifier Based on Gaussian Mixture Model in the Concept Drift Situation

Dimensionality reduction in the context of dynamic social media data streams

Passive concept drift handling via variations of learning vector quantization

Contact Info

Product

Resources

About