• Corpus ID: 11771620

m-TSNE: A Framework for Visualizing High-Dimensional Multivariate Time Series

  title={m-TSNE: A Framework for Visualizing High-Dimensional Multivariate Time Series},
  author={Minh Quoc Nguyen and S. Purushotham and Hien To and Cyrus Shahabi},
Multivariate time series (MTS) have become increasingly common in healthcare domains where human vital signs and laboratory results are collected for predictive diagnosis. Recently, there have been increasing efforts to visualize healthcare MTS data based on star charts or parallel coordinates. However, such techniques might not be ideal for visualizing a large MTS dataset, since it is difficult to obtain insights or interpretations due to the inherent high dimensionality of MTS. In this paper… 

Figures from this paper

High-Dimensional Multivariate Time Series Data Visualization Using an Open Source Framework

The paper disscuses the features and various modules of Visbrain for current understanding research trends in the field of Visual Analytics.

Visualizing Time Series Data with Temporal Matching Based t-SNE

Two temporal matching metrics are proposed, namely, dynamic time warping (DTW) and angular metric for shape similarity (AMSS), which provide a more robust similarity metric for time series data so that the embedding process in t-SNE can be made more effective.

Quantitative Evaluation of Time‐Dependent Multidimensional Projection Techniques

An approach to assess projection techniques for dynamic data and understand the relationship between visual quality and stability, and elect the best‐suited methods for projecting time‐dependent multivariate data.

Visualizing and Exploring Dynamic High-Dimensional Datasets with LION-tSNE

LION-tSNE (Local Interpolation with Outlier coNtrol) - a novel approach for incorporating new data into tSNE representation based on local interpolation in the vicinity of training data, outlier detection and a special outlier mapping algorithm is proposed.

Getting over High-Dimensionality: How Multidimensional Projection Methods Can Assist Data Science

This research presents a comprehensive overview of current state-of-the-art multidimensional projection techniques and shows example codes in Python and R languages, all available on the internet.

Dimensionality Reduction for Visualization of Time Series and Trajectories

This work has shown that the applicability of approaches to transform multivariate time series data to one-dimensional data by reducing the number of features in order to visualize their time-dependent behaviors is restricted to a limited number of data instances that can be visualized simultaneously.

Ant-SNE: Tracking Community Evolution via Animated t-SNE

We introduce a method for tracking the community evolution and a prototype (Ant-SNE) for analyzing multivariate time series and guiding interactive exploration through high-dimensional data. The

Visualising deep network time-series representations

A method that addresses the issue of viewing the causal link between the input and output of the model without presenting how the model learns to represent the data is proposed, with a focus on visualising multi-dimensional time-series data.

k-NN Sampling for Visualization of Dynamic Data Using LION-tSNE

Two new sampling methods are proposed which are based on k-NN (k-Nearest Neighbor) graph update properties which outperform existing LION-tSNE method with 0.5 to 2% more k-nn accuracy and results are more consistent.

DeepVix: Explaining Long Short-Term Memory Network With High Dimensional Time Series Data

This paper aims to combine the strengths of both data science fields into a unified system, called DeepVix, which focuses on the visual explainability of the multivariate time-series predictions using neural networks.



Temporal Event Sequence Simplification

This paper presents a series of user-driven data simplifications that allow researchers to pare event records down to their core elements, and presents a novel metric for measuring visual complexity, and a language for codifying disjoint strategies into an overarching simplification framework.

Interactive Information Visualization to Explore and Query Electronic Health Records

This monograph is written for both scientific researchers and designers of future user interfaces for EHRs to help them understand this vital domain and appreciate the features and virtues of existing systems, so they can create still more advanced systems.

Personalized Mortality Prediction Driven by Electronic Medical Data and a Patient Similarity Metric

It is proved that analyzing only similar patients leads to better outcome prediction performance than analyzing all available patients, and the trade-off between training data size and the degree of similarity between the training data and the index patient for whom prediction is to be made.

Outflow : Visualizing Patient Flow by Symptoms and Outcome

The Outflow visualization technique is described, designed to summarize temporal event data that has been extracted from the EMRs of a cohort of patients, to help clinicians understand how certain progression paths may lead to better or worse outcomes.

Predicting Patient's Trajectory of Physiological Data using Temporal Trends in Similar Patients: A System for Near-Term Prognostics.

A novel system is presented, which leverages inter-patient similarity for retrieving patients who display similar trends in their physiological time-series data, and which is used to project patient data into the future to provide insights for the query patient.

Connecting time-oriented data and information to a coherent interactive visualization

Different interactive visualization techniques are presented which enable the users to reveal the data at several levels of detail and abstraction, ranging from a broad overview to the fine structure, and a time visualization and navigation technique that connects overview+detail, pan+zoom, and focus+context features to one powerful time-browser.

A PCA-based similarity measure for multivariate time series

The results show the superiority of the approaches as compared to the traditional similarity measures for MTS datasets, such as Euclidean Distance (ED), Dynamic Time Warping (DTW), Weighted Sum SVD (WSSVD) and PCA similarity factor (S<sc>PCA</sc>) in precision/recall.

DecisionFlow: Visual Analytics for High-Dimensional Temporal Event Sequence Data

The study results demonstrate that DecisionFlow enables the quick and accurate completion of a range of sequence analysis tasks for datasets containing thousands of event types and millions of individual events.