Advances in software engineering, particularly in Agile software development (ASD), demand innovative approaches to effort estimation due to the volatility in Agile environments. Recent trends have made the automation of story point (SP) estimation increasingly relevant, with significant potential for enhancing accuracy. This study introduces a novel model for software effort estimation (SEE) utilizing a deep learning (DL)-based sentence-BERT (SBERT) model for feature extraction combined with advanced gradient-boosted tree (GBT) algorithms. A comprehensive evaluation shows that the proposed model outperforms standard SEE and state-of-the-art models, demonstrating a mean absolute error (MAE) of 2.15 and a median absolute error (MdAE) of 1.85, representing a 12% improvement over the baseline model and an 18% improvement over the best-performing state-of-the-art model. The standardized accuracy (SA) is 93%, which is 7% higher than the next best model, across a dataset of 31,960 issues from 26 open-source Agile projects. This study contributes to software engineering by offering a more accurate and reliable decision support system for estimating project efforts.
1 April 2020 in International Journal of Cognitive Informatics and Natural Intelligence
The possibility to enhance prediction accuracy for foreign exchange rates was investigated in two ways: first applying an outside the box approach to modeling price graphs by exploiting their visual properties, and secondly employing the most efficient methods to detect patterns to classify the direction of movement. The approach that exploits the visual properties of price graphs which make use of density regions along with high and low values describing the shape; hence, the authors propose the name ‘Finance Vision.' The data used in the predictive model consists of 1-hour past price values of 4 different currency pairs, between 2003 and 2016. Prediction performances of state-of-the-art methods; Extreme Gradient Boosting, Artificial Neural Network and Support Vector Machines are compared over the same data with the same sets of features. Results show that density based visual features contribute considerably to prediction performance.
Background/Objectives: Breast ultrasound is widely used for the diagnostic evaluation of breast lesions; however, reliable lesion characterization remains challenging due to substantial image heterogeneity and the limited size of most clinically available datasets. These constraints reduce the generalizability of end-to-end deep learning approaches in routine practice. The objective of this study was to evaluate a data-efficient diagnostic framework that integrates image-derived features with clinical sonographic descriptors to improve breast ultrasound lesion classification in small cohorts. Methods: Ultrasound images from the publicly available BrEaST-Lesions dataset were processed using a pretrained convolutional neural network to extract compact image feature representations from full images, lesion masks, and cropped tumor regions. These features were combined with manually recorded sonographic descriptors after label encoding to form a unified tabular dataset. Gradient-boosted tree models were trained using descriptor-only and fused feature sets with fivefold stratified cross-validation and evaluated on an independent external hold-out test set. Results: Using sonographic descriptors alone, the best-performing model (LightGBM) achieved an external validation accuracy of 0.88, with an area under the receiver operating characteristic curve (AUC) of 0.95. Incorporation of image-derived features improved diagnostic performance on the external test set, yielding an accuracy of 0.88, an AUC of 0.96, and a sensitivity of 1.00 for malignant lesion detection. The fused framework demonstrated more stable generalization than descriptor-only models, particularly for malignant cases. Conclusions: Combining image-derived features with clinical sonographic descriptors within a tabular learning framework provides a robust and data-efficient approach for breast ultrasound-based lesion classification. This strategy supports diagnostic decision-making in small ultrasound datasets and represents a clinically realistic alternative when large-scale deep learning models are impractical.
Background/Objectives: Survival prediction in non-small cell lung cancer (NSCLC) remains challenging, particularly in limited-sample settings where end-to-end deep learning models may suffer from limited generalization. This study aimed to develop a data-efficient, multimodal, and explainable framework integrating computed tomography (CT)-derived imaging information with clinical variables for NSCLC survival prediction. Methods: CT images, tumor segmentations, and clinical data from the publicly available NSCLC Radiomics (LUNG1) dataset (377 patients) were used. Tumor-focused regions were extracted using segmentation masks, and pretrained RadImageNet-InceptionV3 embeddings were obtained from the largest tumor-containing slice and neighboring-slice summaries. Deep imaging embeddings, engineered imaging features, and clinical variables were fused into a unified tabular representation. To improve robustness under limited-sample conditions, feature blocks were compressed using principal component analysis. CatBoost, XGBoost, and LightGBM models were trained on a development set and evaluated on a strictly held-out final validation set. Results: In three-class survival stratification, assigning censored/non-event patients to the upper survival group produced the strongest ordinal prognostic performance. Under the EX_PLUS_NON_EX_TOP setting, CatBoost achieved the best holdout score-based class C-index of 0.655. In continuous survival regression, LightGBM achieved the best holdout event-patient C-index of 0.576. Clinical variables provided the dominant prognostic signal, while compact deep image embeddings contributed complementary information, particularly in separating short- and long-survival groups. SHAP analysis confirmed contributions from both clinical and image-derived features. Conclusions: The proposed framework provides a proof-of-concept demonstration of a data-efficient and explainable image-to-tabular approach for NSCLC survival prediction under strict internal holdout validation. The results suggest that pretrained CT embeddings, clinical variables, gradient-boosted trees, and SHAP-based interpretation can be combined in a feasible, limited-sample survival modeling pipeline, while external validation remains necessary before clinical translation.