Yazar "Tunali, Volkan" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim Yok
    Yayın
    Analysis of function-call graphs of open-source software systems using complex network analysis
    (PAMUKKALE UNIV, 2020) Tunali, Volkan; Tuysuz, Mehmet Ali Aksoy
    Software systems are usually designed in a modular and hierarchical fashion, where functional responsibility of a system is decomposed into multiple functional software elements optimally such as subsystems, modules, packages, classes, methods, and functions. These elements are coupled with each other with some kind of dependency relationships to some degree, and their interactions naturally form a graph or network structure. In this study, we generated the static function-call graphs of several open-source software systems, where functions were the most basic type of interacting elements calling each other. Then, we analyzed the call graphs both visually and topologically using the techniques of complex network analysis. We found the call graphs to reveal scale-free and small-world network properties similar to the findings of the previous studies. In addition, we identified the most central and important functions in each call-graph using several centrality measures. We also performed community analysis and found that the call graphs exhibited a tendency to form communities. Finally, we showed that analysis of static function-call graphs of software systems through complex network analysis has the potential to reveal useful information about them.
  • Küçük Resim Yok
    Yayın
    An Improved Clustering Algorithm for Text Mining: Multi-Cluster Spherical K-Means
    (ZARKA PRIVATE UNIV, 2016) Tunali, Volkan; Bilgin, Turgay; Camurcu, Ali
    Thanks to advances in information and communication technologies, there is a prominent increase in the amount of information produced specifically in the form of text documents. In order to, effectively deal with this "information explosion" problem and utilize the huge amount of text databases, efficient and scalable tools and techniques are indispensable. In this study, text clustering which is one of the most important techniques of text mining that aims at extracting useful information by processing data in textual form is addressed. An improved variant of spherical K-Means (SKM) algorithm named multi-cluster SKM is developed for clustering high dimensional document collections with high performance and efficiency. Experiments were performed on several document data sets and it is shown that the new algorithm provides significant increase in clustering quality without causing considerable difference in CPU time usage when compared to SKM algorithm.