DCEnt‐PredictiveNet: a novel explainable hybrid model for time series forecasting

This work presents a novel hybrid framework called DCEnt-PredictiveNet (deep convolutional neural network (DCNN) + entropy + support vector regressor (SVR)) that concatenate both deep and handcrafted features for time series data analysis and forecasting. From the discrete wavelet transform coeffici...

Full description

Saved in:
Bibliographic Details
Main Authors: Sudarshan, Vidya K., Ramachandra, Reshma A., Ojha, Smit, Tan, Ru-San
Other Authors: College of Computing and Data Science
Format: Article
Language:English
Published: 2024
Subjects:
Online Access:https://hdl.handle.net/10356/180716
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-180716
record_format dspace
spelling sg-ntu-dr.10356-1807162024-10-22T00:36:30Z DCEnt‐PredictiveNet: a novel explainable hybrid model for time series forecasting Sudarshan, Vidya K. Ramachandra, Reshma A. Ojha, Smit Tan, Ru-San College of Computing and Data Science Computer and Information Science Time series forecasting COVID prediction This work presents a novel hybrid framework called DCEnt-PredictiveNet (deep convolutional neural network (DCNN) + entropy + support vector regressor (SVR)) that concatenate both deep and handcrafted features for time series data analysis and forecasting. From the discrete wavelet transform coefficients of input time series data, computed four different handcrafted entropy features, which were then concatenated with deep features extracted using a modified DCNN. The concatenated deep and handcrafted feature vector was then fed to a SVR for prediction. The DCEnt-PredictiveNet framework was trained and tested on three time series datasets of real-world COVID-19, stock price and traffic information, and achieved mean absolute percentage errors of 0.03 %, 1.53 % and 11.41 % for daily cumulative COVID-19 positive cases, closing stock price, and hourly traffic (vehicle numbers) at one junction predictions, respectively. In addition, we incorporated local interpretable model-agnostic explanations and Shapley additive explanations methods into DCEnt-PredictiveNet to enable visualization of significant features that contributed to the model's decision-making, thereby enhancing its explainability. Our DCEnt-PredictiveNet model yielded promising and interpretable forecasting results, which can facilitate advance resource planning in hospitals for incoming COVID-19 patients, stock market investment planning, and efficient traffic control management. 2024-10-22T00:36:30Z 2024-10-22T00:36:30Z 2024 Journal Article Sudarshan, V. K., Ramachandra, R. A., Ojha, S. & Tan, R. (2024). DCEnt‐PredictiveNet: a novel explainable hybrid model for time series forecasting. Neurocomputing, 608, 128389-. https://dx.doi.org/10.1016/j.neucom.2024.128389 0925-2312 https://hdl.handle.net/10356/180716 10.1016/j.neucom.2024.128389 2-s2.0-85202200005 608 128389 en Neurocomputing © 2024 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Computer and Information Science
Time series forecasting
COVID prediction
spellingShingle Computer and Information Science
Time series forecasting
COVID prediction
Sudarshan, Vidya K.
Ramachandra, Reshma A.
Ojha, Smit
Tan, Ru-San
DCEnt‐PredictiveNet: a novel explainable hybrid model for time series forecasting
description This work presents a novel hybrid framework called DCEnt-PredictiveNet (deep convolutional neural network (DCNN) + entropy + support vector regressor (SVR)) that concatenate both deep and handcrafted features for time series data analysis and forecasting. From the discrete wavelet transform coefficients of input time series data, computed four different handcrafted entropy features, which were then concatenated with deep features extracted using a modified DCNN. The concatenated deep and handcrafted feature vector was then fed to a SVR for prediction. The DCEnt-PredictiveNet framework was trained and tested on three time series datasets of real-world COVID-19, stock price and traffic information, and achieved mean absolute percentage errors of 0.03 %, 1.53 % and 11.41 % for daily cumulative COVID-19 positive cases, closing stock price, and hourly traffic (vehicle numbers) at one junction predictions, respectively. In addition, we incorporated local interpretable model-agnostic explanations and Shapley additive explanations methods into DCEnt-PredictiveNet to enable visualization of significant features that contributed to the model's decision-making, thereby enhancing its explainability. Our DCEnt-PredictiveNet model yielded promising and interpretable forecasting results, which can facilitate advance resource planning in hospitals for incoming COVID-19 patients, stock market investment planning, and efficient traffic control management.
author2 College of Computing and Data Science
author_facet College of Computing and Data Science
Sudarshan, Vidya K.
Ramachandra, Reshma A.
Ojha, Smit
Tan, Ru-San
format Article
author Sudarshan, Vidya K.
Ramachandra, Reshma A.
Ojha, Smit
Tan, Ru-San
author_sort Sudarshan, Vidya K.
title DCEnt‐PredictiveNet: a novel explainable hybrid model for time series forecasting
title_short DCEnt‐PredictiveNet: a novel explainable hybrid model for time series forecasting
title_full DCEnt‐PredictiveNet: a novel explainable hybrid model for time series forecasting
title_fullStr DCEnt‐PredictiveNet: a novel explainable hybrid model for time series forecasting
title_full_unstemmed DCEnt‐PredictiveNet: a novel explainable hybrid model for time series forecasting
title_sort dcent‐predictivenet: a novel explainable hybrid model for time series forecasting
publishDate 2024
url https://hdl.handle.net/10356/180716
_version_ 1814777814160769024