VixLSTM: Visual Explainable LSTM for Multivariate Time Series

Tommy Dang, Huyen N. Nguyen, Ngan V.T. Nguyen

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Neural networks are known for their predictive capability, leading to vast applications in various domains. However, the explainability of a neural network model remains enigmatic, especially when the model comes short in learning a particular pattern or features. This work introduces a visual explainable LSTM network framework focusing on temporal prediction. The hindrance to the training process is highlighted by the irregular instances throughout the entire architecture, from input to intermediate layers and output. The framework provides interactive features to support users in customizing and rearranging the structure to obtain different network representations and perform what-if analysis. To evaluate the usefulness of our approach, we demonstrate the application of VixLSTM on the various datasets generated from different domains.

Original languageEnglish
Title of host publicationIAIT 2021 - 12th International Conference on Advances in Information Technology
Subtitle of host publicationIntelligence and Innovation for Digital Business and Society
PublisherAssociation for Computing Machinery
ISBN (Electronic)9781450390125
DOIs
StatePublished - Jun 29 2021
Event12th International Conference on Advances in Information Technology: Intelligence and Innovation for Digital Business and Society, IAIT 2021 - Virtual, Online, Thailand
Duration: Jun 29 2021Jul 1 2021

Publication series

NameACM International Conference Proceeding Series

Conference

Conference12th International Conference on Advances in Information Technology: Intelligence and Innovation for Digital Business and Society, IAIT 2021
Country/TerritoryThailand
CityVirtual, Online
Period06/29/2107/1/21

Keywords

  • Deep Learning Scatterplot
  • Neural Networks
  • Time Series Visualization

Fingerprint

Dive into the research topics of 'VixLSTM: Visual Explainable LSTM for Multivariate Time Series'. Together they form a unique fingerprint.

Cite this