Abstract Function Point Analysis (FPA) is a method in software engineering that focuses on identifying the functions provided by a software system to users, such as data input, processing, output, and database management. These functions are classified according to complexity to quantify the system’s size in functional point units. In this paper, we propose two graph neural networks: a Graph-based Similarity Detection Neural Network (GSDNN) and a Prior-Structural Information Graph Neural Network (PSI-GNN) with a pre-trained layer using transfer learning, to define the best model for functional size prediction and uncover patterns and trends in data. Additionally, the NESMA (Netherlands Software Metrics Users Association) method, from the functional families approach, will be in focus, where the ISBSG (International Software Benchmarking Standards Group) dataset, which provides standardized and relevant data for comparing software performance, was used to analyze 1704 industrial software projects. The goal was to identify the graph architecture with the smallest number of experiments to be performed and the lowest Mean Magnitude Relative Error (MMRE) using orthogonal-array tuning optimization via Latin Square extraction. In the proposed approach, the number of experiments is fewer than 8 for each dataset, and a minimum MMRE value of 0.97% was obtained using PSI-GNN. Additionally, the impact of five input features on the change in MMRE value was analyzed with the top-performing model, employing the SHAP (SHapley Additive exPlanations) feature importance method, visualized through GraphExplainer. The frequency of user-initiated transactions, quantified technically, emerged as the most significant determinant within the NESMA framework.

The purpose of this research is to demonstrate the feasibility of automating ‘die-polygon-capturing’, an economical yet still labor-intensive technique for circuit extraction during the reverse-engineering of simple integrated circuits. As microchip designs become increasingly diverse with the ongoing trend of using application-specific integrated circuits, and considering the importance of reverse-engineering these components, die-polygon-capturing is set to play a more critical role in the future. Due to the apparent absence of prior scientific publications and limited automation efforts in this area, this paper presents a proof of concept for automating the die-polygon-capturing technique, thereby addressing a notable gap in the existing literature, with an overarching goal of reducing the labor-intensity of die-polygon-capturing. Our method consists of training deep neural networks on variations of a dataset and evaluating their segmentation capabilities of various layers and connections in an integrated circuit. Given the limited accessibility to high-quality labeled datasets, the dataset used for this research consists of two images of a single microchip, the AMD 9085D. We implemented a data augmentation process that expanded this dataset to as many as 4872 images. The experiment’s results proof the automation’s feasibility, demonstrating high scores on Intersection over Union and F-beta evaluation metrics. However the primary conclusion drawn is the need for focus on generalizability towards other types of microchips in order to effectively automate this technique for circuit extraction. • This research automates the die-polygon-capturing technique as a proof of concept. • Automating circuit layers’ semantic segmentation substantially reduces manual labor. • Each feature-set variant offers unique advantages for predicting specific labels. • Augmentation positioning influences the quality of segmentation across different layers. • Future improvements in generalization can scale automated die-polygon-capturing.

In this paper, our objective is to find a balanced trade-off between interpretability and accurate, reliable Issue classification-specifically Bug severity-by creating an ensemble Machine Learning (ML) and Natural Language Processing (NLP) methodological approach that enhances software maintenance in Agile development environments. Using the TAWOS dataset, we explored the capabilities of state-of-the-art models such as eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Categorical Boosting (CatBoost), integrated with advanced NLP techniques such as Term Frequency-Inverse Document Frequency (TF-IDF) and Singular Value Decomposition (SVD) for feature extraction. Our results demonstrate that CatBoost achieved superior performance, with accuracies of 99. 61% for ’high’ labeled severity bugs, 97.73% for ’Critical’ , and 97.65% for ’Blocker’ . SHapley Additive exPlanations (SHAP) analysis further identified key semantic descriptors, such as ”crash” and ”timeout,” as critical predictors in these models. Moreover, this research addresses a critical gap in software engineering by improving the precision and efficiency of bug triaging processes, thereby supporting more effective resource allocation and reducing costs in Agile software projects. Finally, the comprehensive preprocessing pipeline we developed, including lemmatization, outlier removal, and non-oversampling techniques, was essential in optimizing model performance, offering a robust framework for enhancing software quality assurance.

In our research, we present a data mining model for the early diagnosis of hyperinsulinemia, potentially reducing the risk of diabetes, heart disease, and other chronic conditions. The dataset, gathered from 2019 to 2022 by Serbia's Healthcare Center through an observational cross-sectional study, includes 1008 adolescents. Medical datasets are often highly imbalanced and may contain irrelevant features that hinder predictive performance. To address these challenges in the medical data analysis, we propose a model employing Functional Principal Component Analysis (FPCA), which also accounts for outliers that could otherwise lead to the inclusion of irrelevant features. Unlike standard Principal Component Analysis (PCA), which is sensitive to the initial positions of cluster centers influencing the final outcome, our model integrates FPCA with K-Means clustering to improve the preprocessing stage. Additionally, we have incorporated the post-hoc explanatory method SHAP (SHapley Additive exPlanations) alongside algorithms such as Random Forest, XGBoost, and LightGBM to provide deeper insights into our model, identifying the most contributory features for the development of hyperinsulinemia. Experimental results showed that combining FPCA with K-Means clustering enhances the accuracy of the XGBoost classifier, with this model achieving an accuracy score of 0.99.

Software estimation rates are still suboptimal regarding efficiency, runtime, and the accuracy of model predictions. Graph Neural Networks (GNNs) are complex, yet their precise forecasting reduces the gap between expected and actual software development efforts, thereby minimizing associated risks. However, defining optimal hyperparameter configurations remains a challenge. This paper compares state-of-the-art models such as Long-Short-Term-Memory (LSTM), Graph Gated Neural Networks (GGNN), and Graph Gated Sequence Neural Networks (GGSNN), and conducts experiments with various hyperparameter settings to optimize performance. We also aim to gain the most informative feedback from our models by exploring insights using a post-hoc agnostic method like Shapley Additive Explanations (SHAP). Our findings indicate that the Taguchi orthogonal array optimization method is the most computationally efficient, yielding notably improved performance metrics. This suggests a compromise between computational efficiency and prediction accuracy while still requiring the lowest number of runnings, with an RMSE of 0.9211 and an MAE of 310.4. For the best-performing model, the GGSNN model, within the Constructive Cost Model (COCOMO), Function Point Analysis (FPA), and Use Case Points (UCP) frameworks, applying the SHAP method leads to a more accurate determination of relevance, as evidenced by the norm reduction in activation vectors. The SHAP method stands out by exhibiting the smallest area under the curve and faster convergence, indicating its efficiency in pinpointing concept relevance.

This paper proposes a post-hoc explanation method for computing concept attribution in Fuzzy Cognitive Map (FCM) models used for scenario analysis, based on SHapley Additive exPlanations (SHAP) values. The proposal is inspired by the lack of approaches to exploit the often-claimed intrinsic interpretability of FCM models while considering their dynamic properties. Our method uses the initial activation values of concepts as input features, while the outputs are considered as the hidden states produced by the FCM model during the recurrent reasoning process. Hence, the relevance of neural concepts is computed taking into account the model’s dynamic properties and hidden states, which result from the interaction among the initial conditions, the weight matrix, the activation function, and the selected reasoning rule. The proposed post-hoc method can handle situations where the FCM model might not converge or converge to a unique fixed-point attractor where the final activation values of neural concepts are invariant. The effectiveness of the proposed approach is demonstrated through experiments conducted on real-world case studies.

Abstract Rapid and accurate assessment of software project development using artificial intelligence tools can be essential for success in the software industry. This article has two objectives: to reduce the magnitude relative error (MRE) value in estimating the effort and cost of software development using the proposed artificial neural network architecture based on the Taguchi method and examine the influence of input variables on the change in relative error value. Clustering and fuzzification methods further mitigate the heterogeneous structure of the different project values of the datasets used. Taguchi method contributes to the reduction of the number of iterations by 99%, which achieves a significant reduction in estimation and value of MRE. By monitoring additional criteria, such as prediction, correlation, and comparing two activation functions, such as sigmoid and radial basis function, the proposed model's correctness, reliability, and stability are confirmed. Significantly better results are expected using the sigmoid activation function and a decrease in the value of the mean (MRE).

This paper proposes a new, improved COmmon Software Measurement International Consortium function point (COSMIC FP) method that uses Artificial Neural Network (ANN) architectures based on Taguchi’s Orthogonal Array to estimate software development effort. The minimum magnitude relative error (MRE) to evaluate these architectures considering the cost effect function, the type of data used in the training, testing, and validation of the proposed models, was used. Applying the fuzzification and clustering method to obtain seven different datasets, we would like to achieve excellent reliability and accuracy of the obtained results. Besides examining the influence of four input values, we aim to reduce the risks of potential errors, increase the coverage of a wide range of different projects and increase the efficiency and success of completing many various software projects. The main contributions of our work are as follows: the influence of four input values of the COSMIC FP method on the change of mean (MRE), development of two simple ANN architectures, the attainment of a small number of performed iterations in software effort estimation (less than 7), reduced software effort estimation time, the use of different values of the International Software Benchmarking Standards Group and other datasets used in the experiment.

Accurate assessment of software project development using the proper artificial intelligence tools can be a significant challenge for success in the software industry. This paper aims to minimize the relative error in software estimation using the proposed model of an artificial neural network (ANN) based on Taguchi's orthogonal vector plan. By selecting methods of clustering and fuzzification of different project values within several used datasets such as COCOMO2000, NASA60, and Kemerer15, reducing the number and time of iterations minimizes Mean Magnitude Relative Error (MMRE) and include a wide range of observed data. Additional criteria, such as monitoring prediction, correlation, and comparison with RBF (Radial Basis Function) relative error, were used to confirm that the proposed model gives two to three times better results depending on the observed cluster. Based on the obtained results, the accuracy and reliability of the proposed model for estimating software projects were determined.

Nowadays, companies are investing in brand new software, given that fact they always need help with estimating software development, effort, costs, and the period of time needed for completing the software itself. In this paper, four different architectures of Artificial Neural Networks (ANN), as one of the most desired tools for predicting and estimating effort in software development, were used. This paper aims to determine the convergence rate of each of the proposed ANNs, when obtaining the minimum relative error, first depending on the cost effect function, then on the nature of the data on which the training, testing, and validation is performed. Magnitude relative error (MRE) is calculated based on Taguchi’s orthogonal plans for each of these four proposed ANN architectures. The fuzzification method, five different datasets, the clustering method for input values of each dataset, and prediction were used to achieve the best model for estimation. Based on performed parts of the experiment, it can be concluded that the convergence rate of each proposed architecture depends on the cost effect function and the nature of projects in different datasets. By following the prediction throughout all experimental parts, it can be further confirmed that ANN-L36 gave the best results in this proposed approach. The main advantages of this model are as follows: the number of iterations is less than 10, shortened effort estimation time thanks to convergence rate, simple architecture of each proposed ANN, large coverage of different values of actual project efficiency, and minimal MMRE. This model can also serve as an idea for the construction of a tool that would be able to reliably, efficiently and accurately estimate the effort when developing various software projects.