Computer Science – Software Engineering
Scientific paper
2010-12-28
Computer Science
Software Engineering
3d Artificial Intelligence Tecniques in Software Engineering Workshop,7 October,2010, Larnaca, Cyprus
Scientific paper
Case Based Reasoning and particularly Estimation by Analogy, has been used in a number of problem-solving areas, such as cost estimation. Conventional methods, despite the lack of a sound criterion for choosing nearest projects, were based on estimation using a fixed and predetermined number of neighbors from the entire set of historical instances. This approach puts boundaries to the estimation ability of such algorithms, for they do not take into consideration that every project under estimation is unique and requires different handling. The notion of distributions of distances together with a distance metric for distributions help us to adapt the proposed method (we call it DD-EbA) each time to a specific case that is to be estimated without loosing in prediction power or computational cost. The results of this paper show that the proposed technique achieves the above idea in a very efficient way.
Angelis Lefteris
Kosti Makrina Viola
Mittas Nikolaos
No associations
LandOfFree
DD-EbA: An algorithm for determining the number of neighbors in cost estimation by analogy using distance distributions does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with DD-EbA: An algorithm for determining the number of neighbors in cost estimation by analogy using distance distributions, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and DD-EbA: An algorithm for determining the number of neighbors in cost estimation by analogy using distance distributions will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-428195