Feature Selection for Agile Development Through Data Mining Techniques
Download
Author:
WAQAS JAWAID
Citable URI :
https://vspace.vu.edu.pk/detail.aspx?id=7
Publisher :
Virtual University of Pakistan
Date Issued:
11/25/2014 12:00:00 AM
Abstract
Traditionally software development has been performed by following a phased model. This model is called the waterfall (or traditional) model, in which the software development life cycle is divided into distinct phases i.e. requirements elicitation, software development, testing, and maintenance, which are followed in a defined order. The waterfall model creates heavy documentation, and most often results in huge rework and cost overruns because customer does not get the visibility of the project until it is very late. As an alternative to the documentation driven, heavyweight software development processes, many lightweight methodologies were created by software practitioners, e.g. crystal methodologies, feature driven development (FDD), scrum, extreme programming (XP) etc. The lightweight methodologies are called agile methodologies, and the development that is done following this model is called agile development. The agile methodologies follow the practices that add value to customers, and accept changes at any time during the development. From its inception, the agile development has been observed to be highly successful and it is used in most of the software companies now. However if a company wants to start using the agile development, then it is very difficult for it to choose which agile practices or features it should follow, because there are many practices for agile development, from which the company must choose which ones it would use. In this work the most important success factors will be extracted from the agile development practices that are successful in the software industry (with reference to the software practitioners in Pakistan). Through literature survey, the candidate factors will be selected, and then a survey will be conducted (using online survey forms) from the software practitioners of Pakistan, to obtain data about which agile practices are the most successful practices? The survey results will be analyzed using state of the art data mining techniques (e.g. dimension reduction techniques, clustering techniques, and regression modeling) on multiple dimensions (extent of usage, extent of benefits) to find out which practices are most successful among the software practitioners? The most useful success factors/features of agile development will be extracted on the basis of the obtained results.
URI :
https://vspace.vu.edu.pk/details.aspx?id=7
Citation:
Jawaid, W. (2014). Feature Selection for Agile Development Through Data Mining Techniques. Virtual University of Pakistan, (Lahore, Pakistan).
Version :
Final Version
Terms of Use :
All the material and results are copyright of Virtual University of Pakistan
Detailed Terms :
Journal :
Files in this item |
Name |
Size |
Format |
Fall 2014_CS720_ms110400039.pdf |
1555kb |
pdf |