A Binary Neural Network Framework for Attribute Selection and Prediction

Victoria Jane Hodge; Tom Jackson; Jim Austin

doi:10.5220/0004150705100515

A Binary Neural Network Framework for Attribute Selection and Prediction

Victoria Jane Hodge, Tom Jackson, Jim Austin

Computer Science

Research output: Contribution to conference › Paper › peer-review

Abstract

In this paper, we introduce an implementation of the attribute selection algorithm, Correlation-based Feature Selection (CFS) integrated with our k-nearest neighbour (k-NN) framework. Binary neural networks underpin our k-NN and allow us to create a unified framework for attribute selection, prediction and classification. We apply the framework to a real world application of predicting bus journey times from traffic sensor data and show how attribute selection can both speed our k-NN and increase the prediction accuracy by removing noise and redundant attributes from the data.

Original language	English
Pages	510-515
Number of pages	5
DOIs	https://doi.org/10.5220/0004150705100515
Publication status	Published - 5 Oct 2012
Event	Proceedings of the 4th International Conference on Neural Computation Theory and Applications (NCTA 2012) - Barcelona, Spain Duration: 5 Oct 2012 → 7 Oct 2012

Conference

Conference	Proceedings of the 4th International Conference on Neural Computation Theory and Applications (NCTA 2012)
Country/Territory	Spain
City	Barcelona
Period	5/10/12 → 7/10/12

Keywords

Attribute Selection
Feature Selection
Binary Neural Network
Prediction
k-Nearest Neighbour

Access to Document

10.5220/0004150705100515

HodgeJacksonAustin_NCTAAccepted author manuscript, 386 KB

Cite this

@conference{e4a71cd09de04e83925586724a7113fa,

title = "A Binary Neural Network Framework for Attribute Selection and Prediction",

abstract = "In this paper, we introduce an implementation of the attribute selection algorithm, Correlation-based Feature Selection (CFS) integrated with our k-nearest neighbour (k-NN) framework. Binary neural networks underpin our k-NN and allow us to create a unified framework for attribute selection, prediction and classification. We apply the framework to a real world application of predicting bus journey times from traffic sensor data and show how attribute selection can both speed our k-NN and increase the prediction accuracy by removing noise and redundant attributes from the data.",

keywords = "Attribute Selection, Feature Selection, Binary Neural Network, Prediction, k-Nearest Neighbour",

author = "Hodge, {Victoria Jane} and Tom Jackson and Jim Austin",

year = "2012",

month = oct,

day = "5",

doi = "10.5220/0004150705100515",

language = "English",

pages = "510--515",

note = "Proceedings of the 4th International Conference on Neural Computation Theory and Applications (NCTA 2012) ; Conference date: 05-10-2012 Through 07-10-2012",

}

TY - CONF

T1 - A Binary Neural Network Framework for Attribute Selection and Prediction

AU - Hodge, Victoria Jane

AU - Jackson, Tom

AU - Austin, Jim

PY - 2012/10/5

Y1 - 2012/10/5

N2 - In this paper, we introduce an implementation of the attribute selection algorithm, Correlation-based Feature Selection (CFS) integrated with our k-nearest neighbour (k-NN) framework. Binary neural networks underpin our k-NN and allow us to create a unified framework for attribute selection, prediction and classification. We apply the framework to a real world application of predicting bus journey times from traffic sensor data and show how attribute selection can both speed our k-NN and increase the prediction accuracy by removing noise and redundant attributes from the data.

AB - In this paper, we introduce an implementation of the attribute selection algorithm, Correlation-based Feature Selection (CFS) integrated with our k-nearest neighbour (k-NN) framework. Binary neural networks underpin our k-NN and allow us to create a unified framework for attribute selection, prediction and classification. We apply the framework to a real world application of predicting bus journey times from traffic sensor data and show how attribute selection can both speed our k-NN and increase the prediction accuracy by removing noise and redundant attributes from the data.

KW - Attribute Selection

KW - Feature Selection

KW - Binary Neural Network

KW - Prediction

KW - k-Nearest Neighbour

U2 - 10.5220/0004150705100515

DO - 10.5220/0004150705100515

M3 - Paper

SP - 510

EP - 515

T2 - Proceedings of the 4th International Conference on Neural Computation Theory and Applications (NCTA 2012)

Y2 - 5 October 2012 through 7 October 2012

ER -