Selecting features in origin analysis

Pam Green; Peter C.R. Lane; Austen Rainer; Sven Bodo Scholz

doi:10.1007/978-0-85729-130-1-29

Selecting features in origin analysis

Pam Green, Peter C.R. Lane, Austen Rainer, Sven Bodo Scholz

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

1 Citation (Scopus)

Abstract

When applying a machine-learning approach to develop classifiers in a new domain, an important question is what measurements to take and how they will be used to construct informative features. This paper develops a novel set of machine-learning classifiers for the domain of classifying files taken from software projects; the target classifications are based on origin analysis. Our approach adapts the output of four copy-analysis tools, generating a number of different measurements. By combining the measures and the files on which they operate, a large set of features is generated in a semi-automatic manner. After which, standard attribute selection and classifier training techniques yield a pool of high quality classifiers (accuracy in the range of 90%), and information on the most relevant features.

Original language	English
Title of host publication	Res. and Dev. in Intelligent Syst. XXVII
Subtitle of host publication	Incorporating Applications and Innovations in Intel. Sys. XVIII - AI 2010, 30th SGAI Int. Conf. on Innovative Techniques and Applications of Artificial Intel.
Pages	379-392
Number of pages	14
DOIs	https://doi.org/10.1007/978-0-85729-130-1-29
Publication status	Published - 01 Dec 2011
Externally published	Yes
Event	30th SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2010 - Cambridge, United Kingdom Duration: 14 Dec 2010 → 16 Dec 2010

Conference

Conference	30th SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2010
Country	United Kingdom
City	Cambridge
Period	14/12/2010 → 16/12/2010

ASJC Scopus subject areas

Artificial Intelligence
Information Systems

Access to Document

10.1007/978-0-85729-130-1-29

Fingerprint Dive into the research topics of 'Selecting features in origin analysis'. Together they form a unique fingerprint.

Cite this

@inproceedings{ed2b87929b694a6986b319ccfac34762,

title = "Selecting features in origin analysis",

abstract = "When applying a machine-learning approach to develop classifiers in a new domain, an important question is what measurements to take and how they will be used to construct informative features. This paper develops a novel set of machine-learning classifiers for the domain of classifying files taken from software projects; the target classifications are based on origin analysis. Our approach adapts the output of four copy-analysis tools, generating a number of different measurements. By combining the measures and the files on which they operate, a large set of features is generated in a semi-automatic manner. After which, standard attribute selection and classifier training techniques yield a pool of high quality classifiers (accuracy in the range of 90%), and information on the most relevant features.",

author = "Pam Green and Lane, {Peter C.R.} and Austen Rainer and Scholz, {Sven Bodo}",

year = "2011",

month = dec,

day = "1",

doi = "10.1007/978-0-85729-130-1-29",

language = "English",

isbn = "9780857291295",

pages = "379--392",

booktitle = "Res. and Dev. in Intelligent Syst. XXVII",

note = "30th SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2010 ; Conference date: 14-12-2010 Through 16-12-2010",

}

Green, P, Lane, PCR, Rainer, A & Scholz, SB 2011, Selecting features in origin analysis. in Res. and Dev. in Intelligent Syst. XXVII: Incorporating Applications and Innovations in Intel. Sys. XVIII - AI 2010, 30th SGAI Int. Conf. on Innovative Techniques and Applications of Artificial Intel.. pp. 379-392, 30th SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2010, Cambridge, United Kingdom, 14/12/2010. https://doi.org/10.1007/978-0-85729-130-1-29

Selecting features in origin analysis. / Green, Pam; Lane, Peter C.R.; Rainer, Austen; Scholz, Sven Bodo.

Res. and Dev. in Intelligent Syst. XXVII: Incorporating Applications and Innovations in Intel. Sys. XVIII - AI 2010, 30th SGAI Int. Conf. on Innovative Techniques and Applications of Artificial Intel.. 2011. p. 379-392.

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

TY - GEN

T1 - Selecting features in origin analysis

AU - Green, Pam

AU - Lane, Peter C.R.

AU - Rainer, Austen

AU - Scholz, Sven Bodo

PY - 2011/12/1

Y1 - 2011/12/1

N2 - When applying a machine-learning approach to develop classifiers in a new domain, an important question is what measurements to take and how they will be used to construct informative features. This paper develops a novel set of machine-learning classifiers for the domain of classifying files taken from software projects; the target classifications are based on origin analysis. Our approach adapts the output of four copy-analysis tools, generating a number of different measurements. By combining the measures and the files on which they operate, a large set of features is generated in a semi-automatic manner. After which, standard attribute selection and classifier training techniques yield a pool of high quality classifiers (accuracy in the range of 90%), and information on the most relevant features.

AB - When applying a machine-learning approach to develop classifiers in a new domain, an important question is what measurements to take and how they will be used to construct informative features. This paper develops a novel set of machine-learning classifiers for the domain of classifying files taken from software projects; the target classifications are based on origin analysis. Our approach adapts the output of four copy-analysis tools, generating a number of different measurements. By combining the measures and the files on which they operate, a large set of features is generated in a semi-automatic manner. After which, standard attribute selection and classifier training techniques yield a pool of high quality classifiers (accuracy in the range of 90%), and information on the most relevant features.

UR - http://www.scopus.com/inward/record.url?scp=84865515802&partnerID=8YFLogxK

U2 - 10.1007/978-0-85729-130-1-29

DO - 10.1007/978-0-85729-130-1-29

M3 - Conference contribution

AN - SCOPUS:84865515802

SN - 9780857291295

SP - 379

EP - 392

BT - Res. and Dev. in Intelligent Syst. XXVII

T2 - 30th SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2010

Y2 - 14 December 2010 through 16 December 2010

ER -