WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection
Permanent URI for this collectionhttps://hdl.handle.net/11147/7150
Browse
4 results
Search Results
Article Citation - WoS: 1Citation - Scopus: 1Automating Software Size Measurement With Language Models: Insights From Industrial Case Studies(Elsevier Science Inc, 2026) Unlu, Huseyin; Tenekeci, Samet; Kennouche, Dhia Eddine; Demirors, OnurObjective software size measurement is critical for accurate effort estimation, yet many organizations avoid it due to high costs, required expertise, and time-consuming manual effort. This often leads to vague predictions, poor planning, and project overruns. To address this challenge, we investigate the use of pre-trained language models - BERT and SE-BERT - to automate size measurement based on textual requirements using COSMIC and MicroM methods. We constructed one heterogeneous dataset and two industrial datasets, each manually measured by experienced analysts. Models were evaluated in three settings: (i) generic model evaluation, where the models are trained and tested on heterogeneous data, (ii) internal evaluation, where the models are trained and tested on organization-specific data, and (iii) external evaluation, where generic models were tested on organization-specific data. Results show that organization-specific models significantly outperform generic models, indicating that aligning training data with the target organization's requirement style is critical for accuracy. SE-BERT, a domain-adapted variant of BERT, improves performance, particularly in low-resource settings. These findings highlight the practical potential of tailoring training data for broader adoption and cost-effective software size measurement in industrial contexts.Article Citation - WoS: 1Citation - Scopus: 1Application of a Size Measurement Standard for Data Warehouse Projects(Wiley, 2024) Unlu, Hueseyin; Yueruem, Ozan Rasit; Yildiz, Ali; Demirors, OnurMethodologyIn this research, we conducted a case study to establish a foundation for size measurement and effort estimation in DWH projects. We first applied a productivity-based estimation approach using linear regression with the ISBSG repository to assist organizations without historical data. We then evaluated various machine learning algorithms to improve estimation accuracy. Finally, we tested a combined model that integrates both approaches for estimating effort in external projects.ResultsUsing the ISBSG dataset, linear regression models based on productivity achieved a Mean Magnitude of Relative Error (MMRE) of 0.285. Machine learning algorithms improved accuracy by 22.81%, reducing the MMRE to 0.220. The final model, applied to external projects, yielded MRE values between 0.010 and 0.245.ConclusionThe ISBSG repository is a valuable resource for effort estimation in DWH projects. Combining productivity-based estimation with machine learning enhances accuracy and predictive performance, making it a more reliable approach than traditional models.Conference Object Citation - WoS: 1Citation - Scopus: 3Automated Estimation of Functional Size From Code(IEEE, 2020) Özen, Özgesu; Özsoy, Bora; Aktılav, Busenur; Güleç, Eren Can; Demirörs, OnurDetermination of the size of a software project is challenging as well as crucial for both self-employed software developers and corporate businesses. That's why it is subjected to a lot of academic studies where it is discussed how to determine the size more accurately. Functional Size Measurement (FSM) is one the most popular measurement techniques for a software from the point of the delivered functionality. However, the aspects of know-how, the cost, time, and manual operation creates difficulties to apply FSM techniques. This study aims to solve these issues by automating the measurement process to approximate the functional size of a project using the COSMIC Functional Size Measurement. The end product of this study is called 'Cosmic APP' that utilizes the sequence diagram of a software after reverse engineering it from the given code using a third-party tool called 'SequenceDiagram'. The working principles, the estimation process, and the obtained results of 'Cosmic APP' are described thoroughly in this paper. © 2020 IEEE.Conference Object Citation - WoS: 15Citation - Scopus: 20Measureability of Functional Size in Agile Software Projects: Multiple Case Studies With Cosmic Fsm(IEEE, 2019) Hacaloğlu, Tuna; Demirörs, OnurFunctional size measurement (FSM) has been used in software engineering for decades as a main driver for estimation and significant input for other various project management activities throughout the project life span. To apply FSM accurately at the early stages of software development process, especially for estimation purposes, functional user requirements need to be available in detail as required by the adopted FSM method. However, in agile software development, requirement specifications, in general, are kept minimal. For this reason, the adjustment of the requirements to the necessary granularity level has been articulated as one of the barriers preventing the diffusion of FSM practices among agile teams. In this paper, we take a closer look at this problem in order to investigate the usability of FSM and to reveal FSM related challenges empirically through case studies on real agile projects from different software organizations. This study also provides a snapshot of agile organizations in terms of requirement specification and estimation related practices. © 2019 IEEE.