Detecting and Tracking Concept Class Drift and Emergence in Non-Stationary Fast Data Streams

Parker, Brandon S.; Khan, Latifur

doi:10.1609/aaai.v29i1.9588

Cited by 19 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper aims at building a classification pipeline from evolving data streams. Several different online learning, or stream learning, algorithms have been proposed in the literature to deal with evolving data streams in supervised [53,54], unsupervised [55] and semi-supervised settings [56]. The static data from AMIGOS and from Experiment I was streamed using a library for online learning: river [57].…”

Section: Online Learning and Progressive Validationmentioning

confidence: 99%

Online Learning for Wearable EEG-based Emotion Classification

Moontaha¹,

Schumann²,

Arnrich³

2023

Preprint

View full text Add to dashboard Cite

Emotions are indicators of affective states and play a significant role in human daily life, behavior, and interactions. Giving emotional intelligence to the machines could, for instance, facilitate early detection and prediction of (mental) diseases and symptoms. Electroencephalography (EEG) -based emotion recognition is being widely applied because it measures electrical correlates directly from the brain rather than the indirect measurement of other physiological responses initiated by the brain. The recent development of non-invasive and portable EEG sensors makes it possible to use them in real-time applications. Therefore, this paper presents a real-time emotion classification pipeline, which trains different binary classifiers for the dimensions of Valence and Arousal from an incoming EEG data stream. After achieving a 23.9% (Arousal) and 25.8% (Valence) higher f1-score on the state-of-art AMIGOS dataset, this pipeline was applied to the dataset achieved by an emotion elicitation experimental framework developed within the scope of this thesis. Following two different protocols, 15 participants were recorded using two different consumer-grade EEG devices while watching 16 short emotional videos in a controlled environment. For an immediate label setting, the mean f1-score of 87% and 82% were achieved for Arousal and Valence, respectively. In a live scenario, while continuously being updated on the incoming data stream with delayed labels, the pipeline proved to be fast enough to achieve predictions in real time. However, the significant discrepancy from the readily available labels on the classification scores leads to future work to include more data with frequent delayed labels in the live settings.

show abstract

Section: Online Learning and Progressive Validationmentioning

confidence: 99%

Online Learning for Wearable EEG-based Emotion Classification

Moontaha¹,

Schumann²,

Arnrich³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Approaches that divide data streams into fixed-size chunks, e.g., (Parker and Khan 2015) cannot capture concept drift immediately if the chunk size is too large, or suffer from unnecessary frequent training during stable time period if the chunk size is too small (Bifet and Gavald 2007). Gradual forgetting is also used in the literature, e.g., (Klinkenberg 2004) to address the infinite length problem of data streams.…”

Section: Related Workmentioning

confidence: 99%

“…Infinite length problem of data streams is typically addressed by dividing the stream into fixed-size chunks, e.g., (Parker and Khan 2015) or using gradual forgetting, e.g., (Klinkenberg 2004). Since data streams are evolving in nature, without prior knowledge of the time-scale of change, both strategies suffer from a trade-off between performance and sensitivity.…”

Section: Introductionmentioning

confidence: 99%

SAND: Semi-Supervised Adaptive Novel Class Detection and Classification over Data Stream

Haque

Khan

Baron

2016

AAAI

Self Cite

View full text Add to dashboard Cite

Most approaches to classifying data streams either divide the stream into fixed-size chunks or use gradual forgetting. Due to evolving nature of data streams, finding a proper size or choosing a forgetting rate without prior knowledge about time-scale of change is not a trivial task. These approaches hence suffer from a trade-off between performance and sensitivity. Existing dynamic sliding window based approaches address this problem by tracking changes in classifier error rate, but are supervised in nature. We propose an efficient semi-supervised framework in this paper which uses change detection on classifier confidence to detect concept drifts, and to determine chunk boundaries dynamically. It also addresses concept evolution problem by detecting outliers having strong cohesion among themselves. Experiment results on benchmark and synthetic data sets show effectiveness of the proposed approach.

show abstract

“…The problem of classification under Streaming Emerging New Class (SENC), which aims to maintain the predictive accuracy for identifying the novel class and the known class in the stream, has recently been attracting an increasing amount of attention and effort due to both the significant research challenges (Masud et al 2011;Parker and Khan 2015;Haque, Khan, and Baron 2016) and the immense practical value (Abdallah et al 2016). A common example in industry is the topic categorization in the news stream, where a new topic of news may arise when a new event occurred.…”

Section: Introductionmentioning

confidence: 99%

Streaming Classification with Emerging New Class by Class Matrix Sketching

Zhu

et al. 2017

AAAI

View full text Add to dashboard Cite

Streaming classification with emerging new class is an important problem of great research challenge and practical value. In many real applications, the task often needs to handle large matrices issues such as textual data in the bag-of-words model and large-scale image analysis. However, the methodologies and approaches adopted by the existing solutions, most of which involve massive distance calculation, have so far fallen short of successfully addressing a real-time requested task. In this paper, the proposed method dynamically maintains two low-dimensional matrix sketches to 1) detect emerging new classes; 2) classify known classes; and 3) update the model in the data stream. The update efficiency is superior to the existing methods. The empirical evaluation shows the proposed method not only receives the comparable performance but also strengthens modelling on large-scale data sets.

show abstract

Detecting and Tracking Concept Class Drift and Emergence in Non-Stationary Fast Data Streams

Cited by 19 publications

References 10 publications

Online Learning for Wearable EEG-based Emotion Classification

Online Learning for Wearable EEG-based Emotion Classification

SAND: Semi-Supervised Adaptive Novel Class Detection and Classification over Data Stream

Streaming Classification with Emerging New Class by Class Matrix Sketching

Contact Info

Product

Resources

About