Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.14279/13755
Title: Reliable confidence intervals for software effort estimation
Authors: Andreou, Andreas S. 
Papadopoulos, Harris 
Papatheocharous, Efi 
Major Field of Science: Engineering and Technology
Field Category: Electrical Engineering - Electronic Engineering - Information Engineering
Keywords: Forecasting;Learning systems;Conformal predictor
Issue Date: Apr-2009
Source: 5th IFIP Conference on Artificial Intelligence Applications and Innovations, AIAI 2009; Thessaloniki; Greece; 23 April 2009 through 25 April 2009; Code 100720
Volume: 475
Conference: International Conference on Artificial Intelligence Applications and Innovations 
Abstract: This paper deals with the problem of software effort estimation through the use of a new machine learning technique for producing reliable confidence measures in predictions. More specifically, we propose the use of Conformal Predictors (CPs), a novel type of prediction algorithms, as a means for providing effort estimations for software projects in the form of predictive intervals according to a specified confidence level. Our approach is based on the well-known Ridge Regression technique, but instead of the simple effort estimates produced by the original method, it produces predictive intervals that satisfy a given confidence level. The results obtained using the proposed algorithm on the COCOMO, Desharnais and ISBSG datasets suggest a quite successful performance obtaining reliable predictive intervals which are narrow enough to be useful in practice.
Description: CEUR Workshop Proceedings Volume 475, 2009, Pages 211-220
ISSN: 1613-0073
Type: Conference Papers
Affiliation : University of Cyprus 
Frederick University 
Publication Type: Peer Reviewed
Appears in Collections:Δημοσιεύσεις σε συνέδρια /Conference papers or poster or presentation

CORE Recommender
Show full item record

SCOPUSTM   
Citations 50

22
checked on Nov 6, 2023

Page view(s) 50

279
Last Week
0
Last month
1
checked on Nov 6, 2024

Google ScholarTM

Check


Items in KTISIS are protected by copyright, with all rights reserved, unless otherwise indicated.