Feature weighting for Case-based reasoning software project effort estimation

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Background: Software effort estimation is one of the most important activities in software development process. Un-fortunately, estimates are often substantially wrong and specifically most projects encounter effort overruns. Numerous methods have been proposed including Case based reasoning (CBR). Existing research shows that feature subset selection (FSS) is an important aspect of CBR, however, searching for the optimal feature weights is a combinatorial problem and therefore NP-hard. Objective: To develop and evaluate efficient algorithms to generalise FSS into an effective feature weighting approach that can improve accuracy further, since not all features contribute equally to solving the problem. Method: Use various search algorithms e.g., forward sequential weighting (FSW) and random mutation hill climbing (RMHC) to assign weight to features in order to improve the estimation accuracy. We will extend an existing CBR java shell ArchANGEL1. We will perform experiments based on repeated measures design on real world datasets to evaluate these algorithms. Limitations of the proposed research: Dataset quality cannot be assured therefore our findings could be influenced by noisy data. Older datasets may be misrepresenting current software development approaches and technologies. CBR could be sensitive to the choice of distance metric; however, we will only use standardised Euclidean distance.

Original languageEnglish
Title of host publication18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014
PublisherAssociation for Computing Machinery
ISBN (Print)9781450324762
DOIs
Publication statusPublished - 2014
Event18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014 - London, United Kingdom
Duration: May 12 2014May 14 2014

Other

Other18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014
CountryUnited Kingdom
CityLondon
Period5/12/145/14/14

Fingerprint

Case based reasoning
Software engineering
Experiments

All Science Journal Classification (ASJC) codes

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Sigweni, B. (2014). Feature weighting for Case-based reasoning software project effort estimation. In 18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014 [a54] Association for Computing Machinery. https://doi.org/10.1145/2601248.2613081
Sigweni, Boyce. / Feature weighting for Case-based reasoning software project effort estimation. 18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014. Association for Computing Machinery, 2014.
@inproceedings{596652d0732c4714a7c86c7a203efee3,
title = "Feature weighting for Case-based reasoning software project effort estimation",
abstract = "Background: Software effort estimation is one of the most important activities in software development process. Un-fortunately, estimates are often substantially wrong and specifically most projects encounter effort overruns. Numerous methods have been proposed including Case based reasoning (CBR). Existing research shows that feature subset selection (FSS) is an important aspect of CBR, however, searching for the optimal feature weights is a combinatorial problem and therefore NP-hard. Objective: To develop and evaluate efficient algorithms to generalise FSS into an effective feature weighting approach that can improve accuracy further, since not all features contribute equally to solving the problem. Method: Use various search algorithms e.g., forward sequential weighting (FSW) and random mutation hill climbing (RMHC) to assign weight to features in order to improve the estimation accuracy. We will extend an existing CBR java shell ArchANGEL1. We will perform experiments based on repeated measures design on real world datasets to evaluate these algorithms. Limitations of the proposed research: Dataset quality cannot be assured therefore our findings could be influenced by noisy data. Older datasets may be misrepresenting current software development approaches and technologies. CBR could be sensitive to the choice of distance metric; however, we will only use standardised Euclidean distance.",
author = "Boyce Sigweni",
year = "2014",
doi = "10.1145/2601248.2613081",
language = "English",
isbn = "9781450324762",
booktitle = "18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014",
publisher = "Association for Computing Machinery",

}

Sigweni, B 2014, Feature weighting for Case-based reasoning software project effort estimation. in 18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014., a54, Association for Computing Machinery, 18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014, London, United Kingdom, 5/12/14. https://doi.org/10.1145/2601248.2613081

Feature weighting for Case-based reasoning software project effort estimation. / Sigweni, Boyce.

18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014. Association for Computing Machinery, 2014. a54.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Feature weighting for Case-based reasoning software project effort estimation

AU - Sigweni, Boyce

PY - 2014

Y1 - 2014

N2 - Background: Software effort estimation is one of the most important activities in software development process. Un-fortunately, estimates are often substantially wrong and specifically most projects encounter effort overruns. Numerous methods have been proposed including Case based reasoning (CBR). Existing research shows that feature subset selection (FSS) is an important aspect of CBR, however, searching for the optimal feature weights is a combinatorial problem and therefore NP-hard. Objective: To develop and evaluate efficient algorithms to generalise FSS into an effective feature weighting approach that can improve accuracy further, since not all features contribute equally to solving the problem. Method: Use various search algorithms e.g., forward sequential weighting (FSW) and random mutation hill climbing (RMHC) to assign weight to features in order to improve the estimation accuracy. We will extend an existing CBR java shell ArchANGEL1. We will perform experiments based on repeated measures design on real world datasets to evaluate these algorithms. Limitations of the proposed research: Dataset quality cannot be assured therefore our findings could be influenced by noisy data. Older datasets may be misrepresenting current software development approaches and technologies. CBR could be sensitive to the choice of distance metric; however, we will only use standardised Euclidean distance.

AB - Background: Software effort estimation is one of the most important activities in software development process. Un-fortunately, estimates are often substantially wrong and specifically most projects encounter effort overruns. Numerous methods have been proposed including Case based reasoning (CBR). Existing research shows that feature subset selection (FSS) is an important aspect of CBR, however, searching for the optimal feature weights is a combinatorial problem and therefore NP-hard. Objective: To develop and evaluate efficient algorithms to generalise FSS into an effective feature weighting approach that can improve accuracy further, since not all features contribute equally to solving the problem. Method: Use various search algorithms e.g., forward sequential weighting (FSW) and random mutation hill climbing (RMHC) to assign weight to features in order to improve the estimation accuracy. We will extend an existing CBR java shell ArchANGEL1. We will perform experiments based on repeated measures design on real world datasets to evaluate these algorithms. Limitations of the proposed research: Dataset quality cannot be assured therefore our findings could be influenced by noisy data. Older datasets may be misrepresenting current software development approaches and technologies. CBR could be sensitive to the choice of distance metric; however, we will only use standardised Euclidean distance.

UR - http://www.scopus.com/inward/record.url?scp=84905484617&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84905484617&partnerID=8YFLogxK

U2 - 10.1145/2601248.2613081

DO - 10.1145/2601248.2613081

M3 - Conference contribution

AN - SCOPUS:84905484617

SN - 9781450324762

BT - 18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014

PB - Association for Computing Machinery

ER -

Sigweni B. Feature weighting for Case-based reasoning software project effort estimation. In 18th International Conference on Evaluation and Assessment in Software Engineering, EASE 2014. Association for Computing Machinery. 2014. a54 https://doi.org/10.1145/2601248.2613081