Search Results (10)

Web service classification plays an important role in software reuse, service discovery, and automatic metadata organization. Although recent deep learning approaches have improved classification performance by using service names and natural-language descriptions, most existing methods still operate as black-box models and offer limited insight into how different metadata sources influence classification decisions. This lack of transparency reduces their practical usefulness for developers who need to verify predicted categories, analyze incorrect classifications, and improve service metadata quality. A well-trained interpretable model can not only help developers choose more appropriate and reliable categories for each web service, but also help write a more reasonable service name and description. In this paper, we present xServeNet, an explainability-oriented extension of ServeNet for transparent web service classification. xServeNet preserves the BERT-based representation and CNN–BiLSTM feature extractor of ServeNet and introduces (i) an instance-wise dynamic source-fusion mechanism that adaptively combines service-name and service-description features according to their semantic contribution, and (ii) model-internal importance indicators at both the source and word levels that support inspection of classification decisions without introducing additional trainable parameters. We benchmark xServeNet against eleven machine learning baselines on two real-world ProgrammableWeb datasets of 10,943 and 14,086 services covering 50 categories. xServeNet reaches 71.08% Top-1/91.35% Top-5 accuracy on the original dataset and 74.10% Top-1/92.95% Top-5 accuracy on the updated dataset, consistently improving Top-1 accuracy over ServeNet while remaining competitive on Top-5, and achieving the lowest per-category Top-5 standard deviation among all twelve compared methods. In practice, the importance indicators support three concrete activities at the service registry: helping developers verify predicted categories at registration time, iterating on description wording when the predicted category looks wrong, and supporting registry curators in flagging likely mislabelled services for review. Full article

Background/Objectives: Obstructive sleep apnea (OSA) and related sleep disorders affect a substantial proportion of hospitalized patients, with an estimated 48% pooled prevalence of undiagnosed OSA in cardiac inpatients and up to 80% of moderate-to-severe community OSA cases carrying no formal diagnosis at the time of hospital admission. In parallel, frailty—a state of heightened physiological vulnerability arising from cumulative multi-system biological decline—is present in 40–80% of inpatients and shares deep, bidirectional neurobiological pathways with sleep-disordered breathing through circadian dysregulation, intermittent hypoxia, hypothalamic–pituitary–adrenal axis activation, and chronic low-grade inflammation. Despite this convergence, no prior study has integrated validated, administratively computable frailty phenotyping with a machine learning framework specifically designed to predict inpatient sleep disorder diagnosis—and OSA in particular—at the point of hospital admission. The present study addresses this gap by developing an admission-time, explainable machine learning framework for the prediction of inpatient sleep disorder diagnoses (ICD-10 G47.x, encompassing OSA G47.3, insomnia G47.0, hypersomnia, and circadian rhythm disorders) and of insomnia specifically (ICD-10 G47.00). Methods: We developed and evaluated a suite of five binary classification models—XGBoost, Random Forest, LightGBM, CatBoost, and Decision Tree—using 9682 balanced hospitalization episodes from the MIMIC-IV (version 2.2) database. The predictor set comprised 23 admission-time structured features across three domains: (i) frailty and comorbidity burden, including the Hospital Frailty Risk Score (HFRS) derived from ICD-10 codes, the Elixhauser comorbidity index, prior admission history, and six binary disease flags (obesity, hypertension, type 2 diabetes, heart failure, COPD, and depression/anxiety); (ii) physiological and laboratory biomarkers from the first 24 h of care, including minimum SpO₂, heart rate variability, hemoglobin, creatinine, albumin, and arterial blood gas parameters; and (iii) sociodemographic and administrative variables encompassing age, sex, ethnicity, insurance type, and admission acuity. Model performance was assessed through five-fold stratified cross-validation and bootstrap confidence intervals (n = 1000 iterations), with predictor importance quantified using SHapley Additive exPlanations (SHAP). Results: XGBoost achieved the strongest aggregate performance across all evaluation metrics, attaining an area under the receiver operating characteristic curve (AUC) of 0.871 (95% CI: 0.856–0.887), accuracy of 79.6%, F1-score of 0.820, and sensitivity of 94.9%, correctly identifying 903 of 952 true positive cases in the held-out test set; all gradient boosting frameworks substantially outperformed the Decision Tree baseline (AUC 0.836). SHAP analysis identified the HFRS and Elixhauser index as the two dominant predictors, followed by depression/anxiety, obesity, hypertension, and minimum SpO₂—a hierarchy that recapitulates the canonical clinical phenotype of obstructive sleep apnea in frail inpatients rather than that of primary insomnia, indicating that the model is preferentially capturing the OSA–frailty axis within the broader G47.x outcome. The predicted probability outputs were well-calibrated across all risk deciles. Conclusions: Frailty-derived features, in combination with admission-time clinical and physiological data, can predict inpatient sleep disorder diagnoses—predominantly OSA—with high sensitivity and well-calibrated risk estimates. The deployable, interpretable nature of the XGBoost model makes it directly suitable for integration into clinical decision support systems, offering a screening tool that requires no dedicated instrumentation beyond routine admission data. By flagging high-risk patients at the moment of admission, the framework provides a concrete mechanism for accelerating referral for definitive diagnostic confirmation (overnight oximetry, polysomnography) and earlier initiation of CPAP and related therapies, with direct implications for reducing the persistent diagnostic gap, perioperative risk, and preventable adverse outcomes in frail hospitalized populations. Full article

Published laboratory data on soil stabilisation are abundant, yet they remain fragmented across studies and are often difficult to reuse because of inconsistent reporting formats, heterogeneous testing conditions, and incomplete metadata. This article presents a curated experimental dataset compiled from 20 published studies on fine-grained soils, comprising 560 records, including 397 unconfined compressive strength (UCS) results and 163 California Bearing Ratio (CBR) results. The dataset is defined by the inclusion of laboratory studies designed around biopolymer-based two-additive stabilisation frameworks, while intentionally retaining untreated and single-additive comparator records reported within the same experimental programmes. This design is a key distinguishing feature of the dataset because it enables analysis of baseline soil behaviour, isolated additive effects, and combined-additive responses within a traceable study context. Across the included studies, the treatment systems cover a wide range of biopolymer- and lignin-related materials, including xanthan gum, guar gum, chitosan, sodium lignosulfonate, and electrolyte lignin stabiliser, together with complementary additives such as cement, lime, fly ash, ground granulated blast-furnace slag, rice husk ash, glass powder, concrete waste, nano-additives, and natural or synthetic fibres. In addition to UCS and CBR outcomes, the dataset preserves key contextual variables required for meaningful secondary reuse, including soil classification, grain-size fractions, Atterberg limits, compaction properties, curing duration, additive identities and dosages, and source-level traceability. The data are distributed as a structured Excel workbook comprising two cleaned outcome-specific sheets (CBR_clean and UCS_clean) and four supporting documentation sheets (StudyInventory, DataDictionary, VocabularyMap, and QC_Log). Record-level identifiers, DOI-linked source fields, inferred-curing flags, and qualified outcome descriptors are retained to support auditability, selective filtering, and reproducible reuse. The resulting dataset provides a practical foundation for comparative assessment of stabilisation strategies, pavement and subgrade engineering studies, meta-analysis, and machine learning applications in geotechnical engineering. Full article

Impact-echo/impact response testing is widely used to detect cracks, voids, and delamination, but transient signals and crowded spectra can complicate diagnosis. This study presents a nonlinear, harmonic-based framework that characterizes delamination using higher-order harmonics in the impact-free response, instead of the amplitude-dependent resonance–frequency shift. The delaminated region is formulated as a locally vibrating nonlinear plate/oscillator with polynomial material and geometric nonlinearities, predicting harmonic components whose levels depend on impact intensity and nonlinearity parameters. The approach is validated on a concrete slab containing an artificial delamination, excited by repeatable impacts, and measured with an accelerometer. Frequency-domain analysis shows that intact regions exhibit a distinct spectral pattern, whereas the delaminated region produces a clear fundamental component and, with modestly increased impacts, a strong second harmonic that serves as a defect signature; time series metrics corroborate nonlinearity. The results demonstrate a nondestructive technique that can localize and characterize delamination without driving the specimen into damaging strain. Looking ahead, the same harmonic signature principle can be extended to vibroacoustic/impact monitoring of lithium-ion batteries to flag mechanically induced internal defects (e.g., separator/electrode delamination) that can precipitate internal short circuits and elevate thermal runaway risk, improving quality control and in-service safety. Full article

A key strategic option for many organizations across the globe is to examine whether and how business ecosystems can help them survive and thrive amidst continuous changes in business realities. Joining a business ecosystem, though, is not a straightforward decision. Current research efforts are falling short of fully identifying a concise and practical set of decision-making factors that potential ecosystem participants can meaningfully use. To address this limitation, the authors developed a framework of decision-making factors (motivations, prerequisites, ecosystem attractiveness), based on (a) their findings of a systematic mapping study they conducted and (b) their parallel research efforts in business ecosystems operations. The proposed framework encompasses a concrete “vocabulary” of decision-making factors that can enable complex “dialogs” between existing and new business ecosystem stakeholders. As a result, this research effort (a) offers a clear and unambiguous categorization of previously overloaded and ambiguous decision-making factors; (b) captures relationships between the three core components of the proposed framework, thus considering upfront any synergies or conflicts among them; and (c) makes the candidate organization’s decision-making process pragmatic, i.e., misalignment among the proposed factors should be considered a ‘red flag’ that may drive the candidate organization to pivot its decision-making process towards another business ecosystem. Full article

Impervious pavement surfaces within urban areas present serious environmental problems due to waterlogging, flooding and in particular, the urban heat island (UHI) phenomenon. Another issue that has recently been highlighted is user comfort in pedestrian and cycling areas. Materials that have potential for overcoming these issues include pervious concrete (PC), a new type of construction material with improved drainage properties and thermal properties. In this study, the thermal properties and behavior of commonly used concrete paving materials in urban areas (dense concrete (DC) and concrete pavers (P)) and pervious concrete (PC) paving flags were investigated and compared in terms of their thermal properties. Material behavior under different temperature conditions was investigated within laboratory research measuring thermal conductivity (λ) and the capacity for heating and cooling using infrared lamp. Complementary to laboratory tests, field research was conducted analyzing the surrounding conditions on pavement wearing course behavior under real weather conditions. Dense concrete paving material had the highest thermal conductivity coefficient and heat absorption capacity, and slowest heating and cooling speed, compared with the other paving materials. The results also highlighted the similar thermal properties of PC and P but with potentially improved user comfort for PC due to its draining properties. The base layer and surrounding characteristics had a significant influence on the thermal behavior of pavements, and future research should consider these parameters when addressing the UHI effect for different paving materials. Full article

The potential to use waste glass, including cathode ray tube (CRT) glass, for making new products or as an admixture to existing ones is being intensively investigated. This kind of research intensified particularly in the period after CRT TV sets and computer monitors were replaced in the market by the advanced technology of thin film transistor (TFT) and liquid crystal display (LCD) screens. Cathode ray tube glass represents a considerable part of electronic waste (e-waste). E-waste globally increases at a far higher rate than other solid waste materials. There is a possibility to recycle cathode ray tube glass and use it in the construction industry. This paper shows the test results of physical and mechanical properties of blocks and paving flags. The reference specimen was made with quartz sand, while the other product employed a combination of quartz sand and ground panel cathode ray tube glass. The glass was ground to the fraction 0.25/1.00 mm, which corresponds to quartz sand fineness. The following tests were performed: shape and dimensions, resistance to freeze/thaw and de-icing salts, water absorption, splitting tensile strength and tensile strength by bending. Special attention was paid to the tests of Böhme wear resistance, slip resistance of the top surface of CRT products using a pendulum, radioactivity and leaching. The texture of the experimental concrete products was observed by SEM (scanning electron microscopy) and analyzed. The results obtained by experimental testing unequivocally show that CRT glass can successfully be used for making concrete blocks and paving flags. Full article

This study focuses on characterisation of side streams including biomass fly ash, biomass bottom ash, coal fly ash, green liquor dregs, limestone mine tailings, and electric arc furnace steel slag from different industrial locations in Finland. It was found that the fly ash samples contained the highest Al₂O₃ and SiO₂ concentrations, a large number of spherical particles of small sizes and high specific surface areas. Fly ashes and steel slag were observed to contain higher amounts of amorphous phases compared to the other side streams. The high loss on ignition value of the coal fly ash and green liquor dregs was found to exceed the limitations for their application in geopolymer composites. Despite their relatively high concentrations in ashes and steel slag, the leaching tests have shown that no hazardous metal leached out from the streams. Finally, test specimens of geopolymer composites (GP2) were prepared by the application of biomass fly ash, bottom ash, and limestone mine tailings without any pre-treatment process, in addition to the ordinary Portland cement-(R) and metakaolin-based geopolymer composites (GP1). The measured compressive (14.1 MPa) and flexural strength (3.5 MPa) of GP2 suggest that it could be used in concrete kerbs and paving flags. The data has also shown that over 500% of the compressive strength was developed between 7 and 28 days in GP2, whereas in the case of reference concrete (R) and the metakaolin-based geopolymer composite (GP1) it was developed in the first 7 days. Full article

This paper aims to assess the influence of clogging on paving material (pervious concrete) drainage characteristics as well as the influence of the properties of an unbound base layer on drainage characteristics of the whole paving system. The clogging influence has been studied measuring the drainage characteristics on pervious concrete flags before and after their clogging, according to ASTM C1701-09. Additionally, the drainage characteristics of uncontaminated pervious concrete as a paving material was assessed using the falling head method. To assess the influence of properties of an unbound base course (UBC) on drainage characteristics of the whole paving system, the unbound base layer was compacted in two different levels of compaction and the drainage characteristics were measured (according to ASTM C1701-09). It is concluded that pervious concrete prepared with a smaller aggregate fraction is more prone to clogging. Regarding the influence of UBC, it is important to find a balance between pervious concrete infiltration and UBC exfiltration rate, particularly in a case of pervious concrete flags made of coarse aggregate. Full article

Shape memory alloys in the form of bars are increasingly used to control structures under seismic loadings. This study investigates the hysteretic behavior and the ultimate energy dissipation capacity of large-diameter NiTi bars subjected to low- and high-cycle fatigue. Several specimens are subjected to quasi-static and to dynamic cyclic loading at different frequencies. The influence of the rate of loading on the shape of the hysteresis loops is analysed in terms of the amount of dissipated energy, equivalent viscous damping, variations of the loading/unloading stresses, and residual deformations. It is found that the log-log scale shows a linear relationship between the number of cycles to failure and the normalized amount of energy dissipated in one cycle, both for low- and for high-cycle fatigue. Based on the experimental results, a numerical model is proposed that consists of two springs with different restoring force characteristics (flag-shape and elastic-perfectly plastic) connected in series. The model can be used to characterize the hysteretic behavior of NiTi bars used as energy dissipation devices in advanced earthquake resistant structures. The model is validated with shake table tests conducted on a reinforced concrete structure equipped with 12.7 mm diameter NiTi bars as energy dissipation devices. Full article