ANFIS and Deep Learning based missing sensor data prediction in IoT

Summary Internet of Things (IoT) consists of billions of devices that generate big data which is characterized by the large volume, velocity, and heterogeneity. In the heterogeneous IoT ecosystem, it is not so surprising that these sensor‐generated data are considered to be noisy, uncertain, erroneo...

Full description

Saved in:
Bibliographic Details
Published in:Concurrency and computation 2020-01, Vol.32 (2), p.n/a
Main Authors: Guzel, Metehan, Kok, Ibrahim, Akay, Diyar, Ozdemir, Suat
Format: Article
Language:English
Subjects:
adaptive‐network based fuzzy inference system (ANFIS)
Deep Learning
Electronic devices
Environment models
Fuzzy systems
Internet of Things
Internet of Things (IoT)
IoT data analysis
Machine learning
Mathematical models
missing sensor data prediction
Parameters
Sensors
Short term
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Summary Internet of Things (IoT) consists of billions of devices that generate big data which is characterized by the large volume, velocity, and heterogeneity. In the heterogeneous IoT ecosystem, it is not so surprising that these sensor‐generated data are considered to be noisy, uncertain, erroneous, and missing due to the lack of battery power, communication errors, and malfunctioning devices. This paper presents Deep Learning (DL) and Adaptive‐Network based Fuzzy Inference System (ANFIS) based prediction models for missing sensor data problem in IoT ecosystem. First, we build ANFIS based models and optimize their parameters. Then, we construct DL based models by using Long Short Term Memory (LSTM) network structure and optimize its parameters by applying the grid search method. Finally, we evaluate all the proposed models with Intel Berkeley Lab dataset. Experimental results demonstrate that the proposed models can significantly improve the prediction accuracy and may be promising for missing sensor data prediction.
ISSN:1532-0626
1532-0634
DOI:10.1002/cpe.5400