Open AccessAnalysis of Software Sizing and Project Estimation prediction by Machine Learning Classification
Author(s) -
A. Sathesh,
Yasir Babiker Hamdan
Publication year - 2022
Publication title -
journal of ubiquitous computing and communication technologies
Language(s) - English
Resource type - Journals
ISSN - 2582-337X
DOI - 10.36548/jucct.2021.4.006
Subject(s) - sizing , estimator , computer science , correctness , software , estimation , process (computing) , software development , machine learning , artificial intelligence , industrial engineering , data mining , statistics , algorithm , mathematics , systems engineering , engineering , art , visual arts , programming language , operating system
In this study, the outcomes of trials with various projects are analyzed in detail. Estimators may decrease mistakes by combining several estimating strategies, which helps them maintain a close eye on the difference between their estimations and reality. An effort estimate is a method for estimating a model's correctness by calculating the total amount of effort needed. It's a major pain in the backside of software development. Several prediction methods have recently been created to find an appropriate estimate. The suggested SVM approach is utilized to reduce the estimation error for the project estimate to the lowest possible value. As a result, throughout the software sizing process, the ideal or exact forecast is achieved. Early in a model's development, the estimate is erroneous since the needs are not defined, but as the model evolves, it becomes more and more accurate. Because of this, it is critical to choose a precise estimate for each software model development. Observations and suggestions for further study of software sizing approaches are also included in the report.