Distributed neural networks for missing big data imputation

Alessio Petrozziello; Ivan Jordanov; Christian Sommeregger

doi:10.1109/IJCNN.2018.8489488

Distributed neural networks for missing big data imputation

Alessio Petrozziello, Ivan Jordanov, Christian Sommeregger

School of Computing

Expedia

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

273 Downloads (Pure)

Abstract

In this paper we investigate the use of Distributed Neural Networks for the imputation of missing values in Big Data context. The presented framework for data imputation is implemented in Spark, allowing easy imputation as an additional step to the data pre-processing pipeline. The Distributed Neural Networks model is using Mini-batch Stochastic Gradient Descent, scaling well with the cluster size and minimizing the communication among the workers. The model is tested on a real-world Recommender Systems dataset, where the missing data is generally a problem for new items, as the systems ranking is usually biased towards the popular items. The model is compared with univariate (Mean and Median Imputation) and multivariate (K-Nearest Neighbours and Linear Regression) imputation techniques, and its performance is validated using prediction accuracy and speed. Furthermore, we evaluate the speedup compared to the sequential implementation of Neural Networks with Stochastic Gradient Descent.

Original language	English
Title of host publication	2018 International Joint Conference on Neural Networks (IJCNN)
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	131-138
Number of pages	9
ISBN (Electronic)	978-1-5090-6014-6
ISBN (Print)	978-1-5090-6015-3
DOIs	https://doi.org/10.1109/IJCNN.2018.8489488
Publication status	Published - 15 Oct 2018
Event	IEEE WCCI 2018, World Congress on Computational Intelligence - Roi de Janeiro, Brazil Duration: 8 Jul 2018 → 13 Jul 2018 http://www.ecomp.poli.br/~wcci2018/

Publication series

Name	IEEE IJCNN Proceedings Series
Publisher	IEEE
ISSN (Electronic)	2161-4407

Conference

Conference	IEEE WCCI 2018, World Congress on Computational Intelligence
Abbreviated title	IJCNN
Country/Territory	Brazil
Period	8/07/18 → 13/07/18
Internet address	http://www.ecomp.poli.br/~wcci2018/

Keywords

Distributed Computation
Big Data
Missing Data Imputation
Neural Networks

Access to Document

10.1109/IJCNN.2018.8489488

IJCNN - Distributed Neural Networks for Missing Big Data Imputation
© © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript (Post-print), 553 KB

Cite this

@inproceedings{6bfbbfcc527f4bcd9b06d1096477a8c1,

title = "Distributed neural networks for missing big data imputation",

abstract = "In this paper we investigate the use of Distributed Neural Networks for the imputation of missing values in Big Data context. The presented framework for data imputation is implemented in Spark, allowing easy imputation as an additional step to the data pre-processing pipeline. The Distributed Neural Networks model is using Mini-batch Stochastic Gradient Descent, scaling well with the cluster size and minimizing the communication among the workers. The model is tested on a real-world Recommender Systems dataset, where the missing data is generally a problem for new items, as the systems ranking is usually biased towards the popular items. The model is compared with univariate (Mean and Median Imputation) and multivariate (K-Nearest Neighbours and Linear Regression) imputation techniques, and its performance is validated using prediction accuracy and speed. Furthermore, we evaluate the speedup compared to the sequential implementation of Neural Networks with Stochastic Gradient Descent.",

keywords = "Distributed Computation, Big Data, Missing Data Imputation, Neural Networks",

author = "Alessio Petrozziello and Ivan Jordanov and Christian Sommeregger",

year = "2018",

month = oct,

day = "15",

doi = "10.1109/IJCNN.2018.8489488",

language = "English",

isbn = "978-1-5090-6015-3",

series = "IEEE IJCNN Proceedings Series",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "131--138",

booktitle = "2018 International Joint Conference on Neural Networks (IJCNN)",

address = "United States",

note = "IEEE WCCI 2018, World Congress on Computational Intelligence, IJCNN ; Conference date: 08-07-2018 Through 13-07-2018",

url = "http://www.ecomp.poli.br/~wcci2018/",

}

Petrozziello, A, Jordanov, I & Sommeregger, C 2018, Distributed neural networks for missing big data imputation. in 2018 International Joint Conference on Neural Networks (IJCNN). IEEE IJCNN Proceedings Series, Institute of Electrical and Electronics Engineers Inc., pp. 131-138, IEEE WCCI 2018, World Congress on Computational Intelligence, Brazil, 8/07/18. https://doi.org/10.1109/IJCNN.2018.8489488

Distributed neural networks for missing big data imputation. / Petrozziello, Alessio; Jordanov, Ivan; Sommeregger, Christian.
2018 International Joint Conference on Neural Networks (IJCNN). Institute of Electrical and Electronics Engineers Inc., 2018. p. 131-138 (IEEE IJCNN Proceedings Series).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Distributed neural networks for missing big data imputation

AU - Petrozziello, Alessio

AU - Jordanov, Ivan

AU - Sommeregger, Christian

PY - 2018/10/15

Y1 - 2018/10/15

N2 - In this paper we investigate the use of Distributed Neural Networks for the imputation of missing values in Big Data context. The presented framework for data imputation is implemented in Spark, allowing easy imputation as an additional step to the data pre-processing pipeline. The Distributed Neural Networks model is using Mini-batch Stochastic Gradient Descent, scaling well with the cluster size and minimizing the communication among the workers. The model is tested on a real-world Recommender Systems dataset, where the missing data is generally a problem for new items, as the systems ranking is usually biased towards the popular items. The model is compared with univariate (Mean and Median Imputation) and multivariate (K-Nearest Neighbours and Linear Regression) imputation techniques, and its performance is validated using prediction accuracy and speed. Furthermore, we evaluate the speedup compared to the sequential implementation of Neural Networks with Stochastic Gradient Descent.

AB - In this paper we investigate the use of Distributed Neural Networks for the imputation of missing values in Big Data context. The presented framework for data imputation is implemented in Spark, allowing easy imputation as an additional step to the data pre-processing pipeline. The Distributed Neural Networks model is using Mini-batch Stochastic Gradient Descent, scaling well with the cluster size and minimizing the communication among the workers. The model is tested on a real-world Recommender Systems dataset, where the missing data is generally a problem for new items, as the systems ranking is usually biased towards the popular items. The model is compared with univariate (Mean and Median Imputation) and multivariate (K-Nearest Neighbours and Linear Regression) imputation techniques, and its performance is validated using prediction accuracy and speed. Furthermore, we evaluate the speedup compared to the sequential implementation of Neural Networks with Stochastic Gradient Descent.

KW - Distributed Computation

KW - Big Data

KW - Missing Data Imputation

KW - Neural Networks

UR - http://www.proceedings.com/41133.html

U2 - 10.1109/IJCNN.2018.8489488

DO - 10.1109/IJCNN.2018.8489488

M3 - Conference contribution

SN - 978-1-5090-6015-3

T3 - IEEE IJCNN Proceedings Series

SP - 131

EP - 138

BT - 2018 International Joint Conference on Neural Networks (IJCNN)

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE WCCI 2018, World Congress on Computational Intelligence

Y2 - 8 July 2018 through 13 July 2018

ER -

Distributed neural networks for missing big data imputation

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this