SeekAView

An intelligent dimensionality reduction strategy for navigating high-dimensional data spaces

Josua Krause, Aritra Dasgupta, Jean Daniel Fekete, Enrico Bertini

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Dealing with the curse of dimensionality is a key challenge in high-dimensional data visualization. We present SeekAView to address three main gaps in the existing research literature. First, automated methods like dimensionality reduction or clustering suffer from a lack of transparency in letting analysts interact with their outputs in real-Time to suit their exploration strategies. The results often suffer from a lack of interpretability, especially for domain experts not trained in statistics and machine learning. Second, exploratory visualization techniques like scatter plots or parallel coordinates suffer from a lack of visual scalability: it is difficult to present a coherent overview of interesting combinations of dimensions. Third, the existing techniques do not provide a flexible workflow that allows for multiple perspectives into the analysis process by automatically detecting and suggesting potentially interesting subspaces. In SeekAView we address these issues using suggestion based visual exploration of interesting patterns for building and refining multidimensional subspaces. Compared to the state-of-The-Art in subspace search and visualization methods, we achieve higher transparency in showing not only the results of the algorithms, but also interesting dimensions calibrated against different metrics. We integrate a visually scalable design space with an iterative workflow guiding the analysts by choosing the starting points and letting them slice and dice through the data to find interesting subspaces and detect correlations, clusters, and outliers. We present two usage scenarios for demonstrating how SeekAView can be applied in real-world data analysis scenarios.

    Original languageEnglish (US)
    Title of host publicationIEEE Symposium on Large Data Analysis and Visualization 2016, LDAV 2016 - Proceedings
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages11-19
    Number of pages9
    ISBN (Electronic)9781509056590
    DOIs
    StatePublished - Mar 8 2017
    Event6th IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2016 - Baltimore, United States
    Duration: Oct 23 2016 → …

    Other

    Other6th IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2016
    CountryUnited States
    CityBaltimore
    Period10/23/16 → …

    Fingerprint

    Dimensionality Reduction
    High-dimensional Data
    Transparency
    Visualization
    Subspace
    Data visualization
    Work Flow
    Refining
    Learning systems
    Scalability
    Statistics
    Scatter diagram
    Dice
    Scenarios
    Curse of Dimensionality
    Data Visualization
    Interpretability
    Slice
    Outlier
    Data analysis

    Keywords

    • Guided Visualization
    • High-Dimensional Data
    • Subspace Exploration

    ASJC Scopus subject areas

    • Computer Science Applications
    • Modeling and Simulation

    Cite this

    Krause, J., Dasgupta, A., Fekete, J. D., & Bertini, E. (2017). SeekAView: An intelligent dimensionality reduction strategy for navigating high-dimensional data spaces. In IEEE Symposium on Large Data Analysis and Visualization 2016, LDAV 2016 - Proceedings (pp. 11-19). [7874305] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/LDAV.2016.7874305

    SeekAView : An intelligent dimensionality reduction strategy for navigating high-dimensional data spaces. / Krause, Josua; Dasgupta, Aritra; Fekete, Jean Daniel; Bertini, Enrico.

    IEEE Symposium on Large Data Analysis and Visualization 2016, LDAV 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 11-19 7874305.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Krause, J, Dasgupta, A, Fekete, JD & Bertini, E 2017, SeekAView: An intelligent dimensionality reduction strategy for navigating high-dimensional data spaces. in IEEE Symposium on Large Data Analysis and Visualization 2016, LDAV 2016 - Proceedings., 7874305, Institute of Electrical and Electronics Engineers Inc., pp. 11-19, 6th IEEE Symposium on Large-Scale Data Analysis and Visualization, LDAV 2016, Baltimore, United States, 10/23/16. https://doi.org/10.1109/LDAV.2016.7874305
    Krause J, Dasgupta A, Fekete JD, Bertini E. SeekAView: An intelligent dimensionality reduction strategy for navigating high-dimensional data spaces. In IEEE Symposium on Large Data Analysis and Visualization 2016, LDAV 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 11-19. 7874305 https://doi.org/10.1109/LDAV.2016.7874305
    Krause, Josua ; Dasgupta, Aritra ; Fekete, Jean Daniel ; Bertini, Enrico. / SeekAView : An intelligent dimensionality reduction strategy for navigating high-dimensional data spaces. IEEE Symposium on Large Data Analysis and Visualization 2016, LDAV 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 11-19
    @inproceedings{9c56b7a84f564149a01265417468b0fb,
    title = "SeekAView: An intelligent dimensionality reduction strategy for navigating high-dimensional data spaces",
    abstract = "Dealing with the curse of dimensionality is a key challenge in high-dimensional data visualization. We present SeekAView to address three main gaps in the existing research literature. First, automated methods like dimensionality reduction or clustering suffer from a lack of transparency in letting analysts interact with their outputs in real-Time to suit their exploration strategies. The results often suffer from a lack of interpretability, especially for domain experts not trained in statistics and machine learning. Second, exploratory visualization techniques like scatter plots or parallel coordinates suffer from a lack of visual scalability: it is difficult to present a coherent overview of interesting combinations of dimensions. Third, the existing techniques do not provide a flexible workflow that allows for multiple perspectives into the analysis process by automatically detecting and suggesting potentially interesting subspaces. In SeekAView we address these issues using suggestion based visual exploration of interesting patterns for building and refining multidimensional subspaces. Compared to the state-of-The-Art in subspace search and visualization methods, we achieve higher transparency in showing not only the results of the algorithms, but also interesting dimensions calibrated against different metrics. We integrate a visually scalable design space with an iterative workflow guiding the analysts by choosing the starting points and letting them slice and dice through the data to find interesting subspaces and detect correlations, clusters, and outliers. We present two usage scenarios for demonstrating how SeekAView can be applied in real-world data analysis scenarios.",
    keywords = "Guided Visualization, High-Dimensional Data, Subspace Exploration",
    author = "Josua Krause and Aritra Dasgupta and Fekete, {Jean Daniel} and Enrico Bertini",
    year = "2017",
    month = "3",
    day = "8",
    doi = "10.1109/LDAV.2016.7874305",
    language = "English (US)",
    pages = "11--19",
    booktitle = "IEEE Symposium on Large Data Analysis and Visualization 2016, LDAV 2016 - Proceedings",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    address = "United States",

    }

    TY - GEN

    T1 - SeekAView

    T2 - An intelligent dimensionality reduction strategy for navigating high-dimensional data spaces

    AU - Krause, Josua

    AU - Dasgupta, Aritra

    AU - Fekete, Jean Daniel

    AU - Bertini, Enrico

    PY - 2017/3/8

    Y1 - 2017/3/8

    N2 - Dealing with the curse of dimensionality is a key challenge in high-dimensional data visualization. We present SeekAView to address three main gaps in the existing research literature. First, automated methods like dimensionality reduction or clustering suffer from a lack of transparency in letting analysts interact with their outputs in real-Time to suit their exploration strategies. The results often suffer from a lack of interpretability, especially for domain experts not trained in statistics and machine learning. Second, exploratory visualization techniques like scatter plots or parallel coordinates suffer from a lack of visual scalability: it is difficult to present a coherent overview of interesting combinations of dimensions. Third, the existing techniques do not provide a flexible workflow that allows for multiple perspectives into the analysis process by automatically detecting and suggesting potentially interesting subspaces. In SeekAView we address these issues using suggestion based visual exploration of interesting patterns for building and refining multidimensional subspaces. Compared to the state-of-The-Art in subspace search and visualization methods, we achieve higher transparency in showing not only the results of the algorithms, but also interesting dimensions calibrated against different metrics. We integrate a visually scalable design space with an iterative workflow guiding the analysts by choosing the starting points and letting them slice and dice through the data to find interesting subspaces and detect correlations, clusters, and outliers. We present two usage scenarios for demonstrating how SeekAView can be applied in real-world data analysis scenarios.

    AB - Dealing with the curse of dimensionality is a key challenge in high-dimensional data visualization. We present SeekAView to address three main gaps in the existing research literature. First, automated methods like dimensionality reduction or clustering suffer from a lack of transparency in letting analysts interact with their outputs in real-Time to suit their exploration strategies. The results often suffer from a lack of interpretability, especially for domain experts not trained in statistics and machine learning. Second, exploratory visualization techniques like scatter plots or parallel coordinates suffer from a lack of visual scalability: it is difficult to present a coherent overview of interesting combinations of dimensions. Third, the existing techniques do not provide a flexible workflow that allows for multiple perspectives into the analysis process by automatically detecting and suggesting potentially interesting subspaces. In SeekAView we address these issues using suggestion based visual exploration of interesting patterns for building and refining multidimensional subspaces. Compared to the state-of-The-Art in subspace search and visualization methods, we achieve higher transparency in showing not only the results of the algorithms, but also interesting dimensions calibrated against different metrics. We integrate a visually scalable design space with an iterative workflow guiding the analysts by choosing the starting points and letting them slice and dice through the data to find interesting subspaces and detect correlations, clusters, and outliers. We present two usage scenarios for demonstrating how SeekAView can be applied in real-world data analysis scenarios.

    KW - Guided Visualization

    KW - High-Dimensional Data

    KW - Subspace Exploration

    UR - http://www.scopus.com/inward/record.url?scp=85017223564&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85017223564&partnerID=8YFLogxK

    U2 - 10.1109/LDAV.2016.7874305

    DO - 10.1109/LDAV.2016.7874305

    M3 - Conference contribution

    SP - 11

    EP - 19

    BT - IEEE Symposium on Large Data Analysis and Visualization 2016, LDAV 2016 - Proceedings

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -