| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

Show document Help

Title:Vpliv zgodovine sprememb vrednosti programskih metrik na napovedovanje trendov vzdrževalnosti odprtokodnih javanskih programskih knjižnic
Authors:ID Gradišnik, Mitja (Author)
ID Karakatič, Sašo (Mentor) More about this mentor... New window
Files:.pdf DOK_Gradisnik_Mitja_2022.pdf (4,16 MB)
MD5: F4843A0F6E64C25B9300EE1009B31C18
 
Language:Slovenian
Work type:Doctoral dissertation
Typology:2.08 - Doctoral Dissertation
Organization:FERI - Faculty of Electrical Engineering and Computer Science
Abstract:Učinkovitost procesa vzdrževanja programskih produktov, ki lahko neposredno vpliva na številne vidike uspešnosti projektov razvoja programskih produktov, vrednotimo skozi različne aspekte, kot so količina vloženega časa, človeški viri ali količina porabljenih finančnih sredstev. Kot ekonomsko učinkoviti veljajo programski produkti, ki jih je mogoče v fazi vzdrževanja hitro in z minimalnim vložkom človeških virov prilagoditi novim potrebam uporabnikov. Sledenje je mogoče enostavno doseči, če so programski produkti grajeni kakovostno. Merjene programskih metrik in vrednotenje notranje strukture daje vpogled dosežene stopnje kakovosti programskega produkta v njegovi opazovani različici. Vrednotenju kakovosti je sicer mogoče dodatno pripeti dimenzijo časa, in sicer tako, da opazujemo njeno spreminjanje tekom evolucije produkta iz različice v različico. Na podlagi vzorcev spreminjanja notranjih atributov izluščenih iz obstoječih programskih produktov je mogoče izgraditi napovedne modele, s katerimi napovemo trende njihovega gibanja v prihodnje. Slednje daje podlago izogibanju pastem, ki jih prinaša slabo vzdrževalna programska oprema. V okviru doktorske disertacije preučujemo programske knjižnice, ki v zadnjem času predstavljajo močan pospeševalnik ponovne uporabe pri razvoju programske opreme. Vodilo predstavljene raziskave tako predstavlja zmožnost uspešnega napovedovanja gibanja vzdrževalnosti programskih knjižnic na podlagi sprememb produktnih programskih metrik predhodnih različic v njihovi sledeči različici. Čedalje boljša sprejetost programskih knjižnic pri razvoju programskih produktov v zadnjem obdobju predstavljata motivacijo, da se v okviru raziskave v ospredje postavi ravno programske knjižnice. Pridobljen vzorec 40 prostodostopnih odprtokodnih javanskih programskih knjižnic nudi osnovo statistični analizi spreminjanja vzdrževalnosti tekom evolucije različic programskih knjižnic. Opravljene analize spreminjanje vzdrževalnosti tekom različic osvetlijo tako na nivoju programskih knjižic kot na nivoju razredov, ki te knjižnice gradijo. Predvsem preučevanje vpliva sprememb vrednosti programskih metrik izmerjenih nad razredi programskih knjižnic predstavlja temelj preučevanja zmožnosti uspešnega napovedovanje trendov gibanja vzdrževalnosti. Ovrednotenje informacijskega doprinosa, ki ga zgodovina sprememb vrednosti programskih metrik doprinese h kakovosti napovednih modelov vzdrževalnosti, predstavlja pomemben element zastavljene raziskave. Doprinos zgodovine sprememb vrednosti programskih metrik napovednim modelov vzdrževalnosti ovrednotimo s pomočjo eksperimenta, v katerem izgradimo in iz vidika kakovosti ocenimo serijo napovednih modelov vzdrževalnosti. Za izgradnjo modelov uporabimo osemnajst različnih razponov zgodovinskih meritev programskih metrik predhodnih različic programskih knjižnic. Pri izbiri pristopov izgradnje napovednih modelov se upiramo na ugotovitve izvedenega sistematičnega pregleda literature.
Keywords:kakovost programske opreme, vzdrževalnost programskih produktov, programske knjižnice, evolucija programskih produktov, programske metrike, napovedni modeli vzdrževalnosti
Place of publishing:Maribor
Publisher:M. Gradišnik
Year of publishing:2022
PID:20.500.12556/DKUM-81563 New window
UDC:004.05:004.428(043.3)
COBISS.SI-ID:125964803 New window
Publication date in DKUM:18.10.2022
Views:568
Downloads:106
Metadata:XML DC-XML DC-RDF
Categories:KTFMB - FERI
:
Copy citation
  
Average score:(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and Share


Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Link:http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:15.04.2022

Secondary language

Language:English
Title:Impact of Historical Software Metric Changes in Predicting Future Maintainability Trends in Open-source Java Software Libraries
Abstract:The efficiency of the software product maintenance process directly affects the success of software product projects, which can be evaluated through several aspects, such as the amount of time invested, human resources or the number of financial resources spent. Software products that can be adapted to new user requirements quickly and with minimal human resources investment are considered cost-effective. Cost-efficiency can be achieved more easily if the software products are built with quality. The measurement of program metrics and the evaluation of the internal structure of software components provide insight into the achieved level of software product quality. The dimension of time can be added to the quality assessment by observing its change during the evolution of the product from version to version. Based on patterns of change of internal attributes extracted from software components, it is possible to build predictive models that can predict the trends of maintainability movement in the future. Efficient maintainability prediction provides us with the basis for avoiding the pitfalls of poorly maintained software. As part of our doctoral dissertation, we study software libraries, recognized recently as a powerful accelerator of reuse in software development. The guideline of the presented research is the ability to successfully predict the movement of maintainability of software libraries in the following version based on changes in product metrics measured in previous versions. The growing acceptance of software libraries in software development in recent years motivates to put software libraries at the forefront of research. The obtained sample of 40 freely available and open-source Java software libraries provides a basis for statistical analysis of changes in maintainability during the evolution of software library components. The presented research shed light on the change in maintenance between the versions, both at the level of software libraries and the level of the classes that build these libraries. Studying the impact of changes in the values of program metrics measured over classes of program libraries is the basis for studying the ability to predict maintainability trends in observed libraries successfully. Evaluating the information contribution that the history of changes in the values of program metrics contributes to the quality of predictive maintenance models is a crucial element of the presented research. The contribution of the history of changes in values of program metrics to predictive models of software maintainability is evaluated using an experiment. As a part of the experiment, we build a series of predictive maintenance models and evaluate them based on their achieved quality. In eighteen iterations of the experiment, we use different ranges of changes in program metrics extracted from previous versions of program libraries to build predictive models. To build predictive models, we rely on the findings of a systematic review of the literature.
Keywords:software quality, software maintainability, software libraries, software evolution, software metrics, maintainability prediction models


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica