Kontaktujte nás | Jazyk: čeština English
Název: | The effects of clustering to software size estimation for the use case points methods |
Autor: | Prokopová, Zdenka; Šilhavý, Radek; Šilhavý, Petr |
Typ dokumentu: | Článek ve sborníku (English) |
Zdrojový dok.: | Advances in Intelligent Systems and Computing. 2017, vol. 575, p. 479-490 |
ISSN: | 2194-5357 (Sherpa/RoMEO, JCR) |
ISBN: | 978-3-319-57140-9 |
DOI: | https://doi.org/10.1007/978-3-319-57141-6_51 |
Abstrakt: | The main objective of the paper is to present the suitability and effects of several different clustering methods for improving accuracy of software size estimation. For software size estimation was used the Algorithmic Optimisation Method (AOM), which is based Use Case Points (UCP) method. The comparison of K-means, Hierarchical and Density-based clustering is provided. Gap, Silhouette and Calinski-Harabasz criterion were selected as an evaluation criterion for clustering quality. Estimation ability of clustered model is compared on Sum of squared error (SSE). Results shows that clustering improves an estimation ability. © Springer International Publishing AG 2017. |
Plný text: | https://link.springer.com/chapter/10.1007/978-3-319-57141-6_51 |
Zobrazit celý záznam |