Further comparison of cross-company and within-company effort estimation models for web applications
Published 23 December 2004
E. Mendes, Barbara Kitchenham
Citations73
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
Abstract
This paper extends a previous study, using data on 67 Web projects from the Tukutuku database, investigating to what extent a cross-company cost model can be successfully employed to estimate effort for projects that belong to a single company, where no projects from this company were used to build the cross-company model. Our within-company model employed data on 14 Web projects from a single Web company. Our results were similar to those from the previous study, showing that predictions based on the within-company model were significantly more accurate than those based on the cross-company model. We also found that predictions were very poor when the within-company cost model was used to estimate effort for 53 Web projects from different companies. We analysed the data using two techniques, forward stepwise regression and case-based reasoning. We found estimates produced using stepwise regression models were better for the within company model while case-based reasoning predictions were better for the cross-company model.
Keywords
Computer Science
Communications of the ACMAn empirical validation of software cost estimation models
903 Citations1987Chris F. Kemerer
Four of the most popular algorithmic models used to estimate software costs (SLIM, COCOMO, Function Points, and ESTIMACS) are evaluated, finding that all of the models tested failed to sufficiently reflect the underlying factors affecting productivity.
IEE Proceedings - SoftwareWhat accuracy statistics really measure
439 Citations2001Barbara Kitchenham, Lesley M. Pickard +2 more
It is demonstrated that MMRE and pred(25) are, respectively, measures of the spread and the kurtosis of the variable z, where z=estimate/actual.
An assessment and comparison of common software cost estimation modeling techniques
275 Citations1999Lionel Briand, Khaled El Emam +3 more
This paper assess and compare a selection of common cost modeling techniques fulfilling a number of important criteria using a large multi-organizational database in the business application domain and shows that the performances of the modeling techniques considered were not significantly different, with the exception of the analogy-based models which appear to be less accurate.
A replicated assessment and comparison of common software cost modeling techniques
220 Citations2000Lionel Briand, Tristen Langley +1 more
Traditional techniques, namely, ordinary least-squares regression and analysis of variance outperformed analogy-based estimation and regression trees and no significant difference was found in accuracy between estimates derived from company-specific data and estimatesderived from multi-organizational data.
Medical Entomology and ZoologyApplied Statistics for Software Managers
216 Citations2002Katrina Maxwell
This work focuses on the development of a Multi-Variable Model for Software Development Productivity and its applications in the context of Software Development Project Data Management.
Using public domain metrics to estimate software development effort
186 Citations2001R. Jeffery, Melanie Ruhe +1 more
The accuracy of cost estimates when applying most commonly used modeling techniques to a large-scale industrial data set which is professionally maintained by the International Software Standards Benchmarking Group (ISBSG) is investigated, and the estimation accuracy is significantly higher than when using the rest of the ISBSG data set.
Information and Software TechnologyA comparative study of two software development cost modeling techniques using multi-organizational and company-specific data
162 Citations2000Richard Jeffery, Martin Ruhé +1 more
Using multi-company data the OLS regression model provided significantly more accurate results than Analogy-based predictions, and it was found in general that models based on the company-specific data resulted in significant more accurate estimates.
Information and Software TechnologyEmpirical studies of assumptions that underlie software cost-estimation models
126 Citations1992BA Kitchenham
The results indicate that the assumption that there is a nonlinear relationship between size and effort is not supported, but the assumption of a non linear relationship between effort and duration is.
IEEE Transactions on Software EngineeringA procedure for analyzing unbalanced datasets
114 Citations1998Barbara Kitchenham
A method of forward pass residual analysis is proposed to analyze unbalanced datasets that include many nominal- and ordinal-scale factors, often found in company datasets used for benchmarking and productivity assessment.
A replicated assessment of the use of adaptation rules to improve Web cost estimation
66 Citations2004E. Mendes, Nile Mosley +1 more
This paper replicates previous work that investigated the use of two types of adaptation rules as a contributing factor to better estimation accuracy, and also investigates theuse of feature subset selection, in addition to adaptation rules.
How valuable is company-specific data compared to multi-company data for software cost estimation?
55 Citations2003Isabella Wieczorek, Melanie Ruhe
A large cost database from Finland that collects standard cost factors and includes information on six individual companies was analyzed and it is shown that company-specific models seem not to yield better results than the company external models.
A comparison of cross-company and within-company effort estimation models for Web applications
49 Citations2004Barbara Kitchenham
It is concluded that cross-company effort estimation model can be useful for companies that do not have past projects from which to develop their own models, however, a cross- company data set should only be used for estimation until it is possible for a Web company to build its own within-company model.
Investigating Early Web Size Measures for Web Cost Estimation
36 Citations2005Emília Mendes, Nile Mosley +1 more
Using simulation to evaluate prediction techniques [for software]
28 Citations2002Martin Shepperd, Gada Kadoda
There are significant differences depending upon the characteristics of the data set and researchers should consider the prediction context when evaluating competing prediction systems, according to the results of a comparison of regression, rule induction and nearest neighbours.