Data Stream Trajectory Analysis Using Sequential Possibilistic Gaussian Mixture Model

Wang, Wenlong; Keller, James M.; Skubic, Marjorie; Popescu, Mihail

doi:10.1109/fuzz-ieee.2019.8858874

Cited by 7 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The motivating streaming algorithm that we use in this paper is a variant of the MU streaming clustering algorithm (MUSC). The version, MUSC II [11], is a modified form of that found in [9] and [10] in which the underlying footprints are modeled by the parameters of the components of a Gaussian Mixture. The novelty of this approach is in the initialization phase and discovery of new structures from the outlier list.…”

Section: Introductionmentioning

confidence: 99%

StreamSoNG: A Soft Streaming Classification Approach

Wang

Keller

Dale

et al. 2022

IEEE Trans. Emerg. Top. Comput. Intell.

Self Cite

View full text Add to dashboard Cite

Examining most streaming clustering algorithms leads to the understanding that they are actually incremental classification models. They model existing and newly discovered structures via summary information that we call footprints. Incoming data is normally assigned crisp labels (into one of the structures) and that structure's footprints are incrementally updated. There is no reason that these assignments need to be crisp. In this paper, we propose a new streaming classification algorithm that uses Neural Gas prototypes as footprints and produces a possibilistic label vector (typicalities) for each incoming vector. These typicalities are generated by a modified possibilistic k-nearest neighbor algorithm. The approach is tested on synthetic and real image datasets with excellent results.

show abstract

Section: Introductionmentioning

confidence: 99%

StreamSoNG: A Soft Streaming Classification Approach

Wang

Keller

Dale

et al. 2022

IEEE Trans. Emerg. Top. Comput. Intell.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The motivating streaming algorithm in this work is a variant of the Missouri University (MU) streaming clustering algorithm (MUSC). Our first proposed streaming algorithm, SPGMM [17], is a modified form of MUSC found in [49] and [50] in which the underlying footprints are modeled by the parameters of the components of a Gaussian…”

Section: Chapter 4 Learning Data Distributions From Streaming Datamentioning

confidence: 99%

“…In this work, I designed two multi-dimensional (multi-D) streaming processing algorithms: (i) Sequential Possibilistic Gaussian Mixture Model (SPGMM) that extends the Gaussian Mixture Model (GMM) into a temporal framework to track the pattern changes in data streams [17] and (ii) Streaming Soft Neural Gas (StreamSoNG) that extends the Neural Gas (NG) and Possibilistic K-Nearest Neighbors (PKNN) into a temporal framework to track the pattern changes in data streams [18]. A major contribution of this work is that trajectory analysis on data streams is defined and conducted upon SPGMM to offer earlier, more sensitive health alerts that are customized to the individual's health trajectory, with fewer false alarms.…”

Section: Chapter 1 Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Early detection of health changes in the elderly using in-home multi-sensor data streams

Wang¹

Self Cite

View full text Add to dashboard Cite

The rapid aging of the population worldwide requires increased attention from health care providers and the entire society. For the elderly to live independently, many health issues related to old age, such as frailty and risk of falling, need increased attention and monitoring. When monitoring daily routines for older adults, it is desirable to detect the early signs of health changes before serious health events, such as hospitalizations, happen, so that timely and adequate preventive care may be provided. By deploying multi-sensor systems in homes of the elderly, we can track trajectories of daily behaviors in a feature space defined using the sensor data. In this work, we investigate a methodology for learning data distribution from streaming data and tracking the evolution of the behavior trajectories over long periods (years) using high dimensional streaming clustering and provide very early indicators of changes in health. If we assume that habitual behaviors correspond to clusters in feature space and diseases produce a change in behavior, albeit not highly specific, tracking trajectory deviations can provide hints of early illness. Retrospectively, we visualize the streaming clustering results and track how the behavior clusters evolve in feature space with the help of two dimension-reduction algorithms, Principal Component Analysis (PCA) and t-distributed Stochastic Neighbor Embedding (t-SNE). Moreover, our tracking algorithm in the original high dimensional feature space generates early health warning alerts if a negative trend is detected in the behavior trajectory. We validated our algorithm on synthetic data, real-world data and tested it on a pilot dataset of four TigerPlace residents monitored with a collection of motion, bed, and depth sensors over ten years. We used the TigerPlace electronic health records (EHR) to understand the residents' behavior patterns and to evaluate and explain the health warnings generated by our algorithm. The results obtained on the TigerPlace dataset show that most of the warnings produced by our algorithm can be linked to health events documented in the EHR, providing strong support for a prospective deployment of the approach.

show abstract